(PDF) “ Therapeutic Metaphors in Engineering: How to cure a building structure” | Ana Roldan-Riejos - Academia.edu

“ Therapeutic Metaphors in Engineering: How to cure a building structure”

Selected Papers from the 2008 Stockholm Metaphor Festival, 2014
Ana Roldan-Riejos
This Paper
A short summary of this paper
37 Full PDFs related to this paper
STOCKHOLM STUDIES IN ENGLISH 105 Selected Papers from the 2008 Stockholm Metaphor Festival Selected Papers from the 2008 Stockholm Metaphor Festival Edited by Nils-Lennart Johannesson and David C. Minugh ©N-L Johannesson, D C Minugh, and Acta Universitatis Stockholmiensis 2013 ISBN electronic version: 978-91-87235-65-8 ISBN printed version: 978-91-87235-66-5 ISSN 0346-6272 Printed in Sweden by US-AB, Stockholm 2014 Distributor: ACTA UNIVERSITATIS STOCKHOLMIENSIS In Memoriam Christina Alm-Arvius (1945–2013) Contents Preface ............................................................................................................ 9 Keynote lectures ........................................................................................... 11 Antonio Barcelona Metonymy Is Not Just A Lexical Phenomenon: On the operation of metonymy in grammar and discourse ........................................................... 13 Gerard Steen When is Metaphor Deliberate? ..................................................................... 47 Session papers ............................................................................................... 65 Ludmilla A’Beckett ACADEMIC LIFE IS WAR: The extended metaphor in Tom Sharpe’s Porterhouse Blue .......................................................................................... 67 Christina Alm-Arvius Iconicity and Poeticity in the Discourse Functions of Figures of Speech ..... 91 Bárbara Eizaga Rebollar Letting the Cat out of the Bag: On idiom use and representation ............... 127 Shelley Ching-yu Hsieh and Elena Kolodkina Emotion Expressed with Eyes and Hands in Three Languages .................. 139 Anna Kryvenko Conventional Animal Metaphors and Cultural Scenarios in the English and Ukrainian Sports-Related Lexicon ....................................................... 159 Ana Roldán-Riejos Therapeutic Metaphors in Engineering: How to cure a building structure.... 173 Rotimi Taiwo Metaphors in Nigerian Political Discourse ................................................. 193 Piotr Twardzisz Metaphors in Commercial Contracts .......................................................... 207 List of contributors...................................................................................... 221 Preface The 2008 Stockholm Metaphor Festival attracted some 30 participants from 15 countries throughout the world: Australia, Belgium, Denmark, France, Hungary, the Netherlands, Nigeria, Norway, Poland, Spain, Sweden, Tai- wan, Ukraine, and the USA. The ten papers included in this volume consist of the two keynote lectures and eight out of the 19 session papers. From the outset, the Metaphor Festival has welcomed contributions dis- cussing metaphor and metonymy as well as other types of figurative lan- guage within different theoretical frameworks and with literary as well as linguistic approaches. The papers included in this collection reflect some of the breadth of the presentations at the 2008 Festival. Of the two keynote speakers, Antonio Barcelona presents an anatomy of metonymy at all linguistic levels, from phonology to discourse, while Gerard Steen explores the conditions required for the use of metaphor to be consid- ered deliberate (as opposed to conventional, automatic and unconscious). In the remaining contributions, A’Beckett studies an extended metaphor running through a Tom Sharpe novel, Alm-Arvius explores iconicity and poeticity in the discourse functions of schemes and tropes, Eizaga Rebollar discusses the representation and storage of conceptual information underly- ing idioms as well as the retrieval and use of idioms in communication, Hsieh and Kolodkina present expressions for hands and eyes as metaphors for emotion in Chinese, Russian and English, Kryvenko investigates animal metaphors in Ukrainian and English sports terminology, Roldán-Riejos charts the role of medical metaphors in engineering discourse, Taiwo traces the use of metaphorical expressions in Nigerian political discourse, and Twardzicz provides a scrutiny of metaphors in commercial contracts. It only remains to thank all the people who have contributed to making the Metaphor Festivals a success so far, both the participants and the admin- istrative staff of the Department of English. On behalf of the organisers of the 2008 Metaphor Festival we would like to thank the Vice-Chancellor of Stockholm University, Professor Kåre Bremer, for generous economic sup- port through the Granholm Foundation. Frescati, December 2011 Christina Alm-Arvius Nils-Lennart Johannesson David C. Minugh Editors 9 Preface to the 2013 edition Christina Alm-Arvius passed away on 1 November, 2013, at the age of 67. We have decided to produce a print version of the current volume, which has been available online since December 2011, as a tribute to Christina’s enthu- isiastic and indefatigable work as a colleague on the Metaphor Festival Committee as well as at the English Department of Stockholm University. She will be sorely missed. Stockholm, December 2013 Nils-Lennart Johannesson David C. Minugh Editors Acknowledgements We are grateful to the Department of English, Stockholm University, for sponsoring the first print run of this volume. We wish to express our thanks to the editors of Stockholm Studies in English for including this volume in the series, and the Board of Acta Uni- versitatis Stockholmiensis for agreeing to publish the volume. The editors 10 Keynote lectures Antonio Barcelona Metonymy Is Not Just A Lexical Phenomenon: On the operation of metonymy in grammar and discourse Abstract. The main goal of the article is to carry out a survey of recent research on the role of conceptual metonymy in grammar and discourse. A related goal is to argue that metonymy is not only (or even mainly) a lexical phenomenon. After de- fining conceptual metonymy from a cognitive-linguistic perspective, the author offers a detailed summary of recent research (including his own contributions) on the functioning of metonymy in conceptualization, phonology, morphology, gram- mar and discourse-pragmatic inferencing. This survey offers evidence that metony- my is a conceptual mechanism (an inferential schema) operating under the lexicon (in phonological categorization and in the meaning and grammatical behavior of certain morphemes), at the lexical level, and above the lexicon (motivating certain aspects of grammar, especially grammatical recategorization, and guiding discourse- pragmatic inferencing, especially indirect speech acts and implicatures). This evi- dence, added to the fact that lexical metonymies are often at the same time grammat- ical and discourse metonymies, supports the claim that metonymy is not only a lexi- cal phenomenon. Keywords: Metonymy, metaphor, conceptualization, grammar, phonology, mor- phology, recategorization, inferencing, indirect speech acts, implicatures 1 Introduction The goal of this article is to present a brief yet fairly detailed survey of re- cent research on the ubiquitous role of metonymy in conceptualization, grammar and discourse.1 A point that will be emphasized throughout the paper is the fact (all too often overlooked, even by some cognitive linguists) that metonymy often operates under (phonology, morphemics) and above (phrases, clauses, sen- 1 The research reported in this paper has been funded in part by a grant awarded to project FFI2008-04585/FILO by the Spanish government, Ministry of Science and Innovation. I am grateful to the editors of this volume, especially Nils-Lennart Johannesson, for their invalua- ble comments and suggestions. 13 tences, utterance and discourse) the lexicon—in other words, that metonymy is not just a lexical phenomenon. Section 2 will be devoted to a very concise presentation of the author’s notion of metonymy. Section 3 is concerned with the important function of metonymy in general conceptualization processes (cognitive models, con- ceptual metaphor, blending) of linguistic (but not necessarily lexical) rele- vance. Section 4 discusses its role in the linguistic units below the lexicon in the traditional linguistic hierarchy, i.e. in phonemic categories and in mor- phemic meaning. Section 5 continues this survey by briefly commenting on lexical metonymies; it will be pointed out there that they are often hard to distinguish from metonymies operating at higher grammatical levels or even at discourse level. Section 6 is fully devoted to the role of metonymy in grammatical constructions and constitutes one of the two main parts of this survey; it deals with the role of metonymy in grammatical recategorization (where it is a major motivational force), in compounding, in phrasal gram- mar, in the English tense-aspect-mood system, in clausal grammar, in anaphora and in grammatical form; a detailed example is provided of its motivational role for a grammatical construction. Section 7 completes the survey by looking at the inference-guiding function of metonymy in dis- course comprehension, with particular attention to indirect speech acts and implicatures; a detailed example regarding the latter is offered at the end of the section. The conclusions are presented in section 8. 2 The notion of metonymy Cognitive linguistics regards metonymy as a cognitive mechanism (Lakoff & Johnson 1980, Lakoff 1987), not merely as a figure of speech, as claimed by traditional rhetoric, or as a mere “contextual effect”, as claimed by relevance theorists (Papafragou 1996). Unfortunately, there is not a uniform notion of metonymy in cognitive linguistics; the term “metonymy” is used in cognitive linguistics to cover some very different phenomena. But all cognitive linguists concur in stress- ing the primarily conceptual nature of metonymy. It would take a long article to discuss at length the particular variants on this basic conceptual approach to metonymy that can be found in the cognitive linguistic literature, and this is not the goal of the present paper. Such a discussion can be found in Barce- lona (2003a); for a volume devoted to the discussion of the notion of meton- ymy, see Benczes, Barcelona & Ruiz de Mendoza (2011). I will simply pre- sent and briefly discuss here my own cognitive-linguistic conception of me- tonymy. This is a broad notion capable of capturing what all of the different phenomena have in common that have been regarded as metonymic in the 14 literature. I call this the schematic notion of metonymy, which contains the necessary and sufficient conditions for metonymicity: Metonymy is an asymmetric mapping of a conceptual entity, the source, onto another conceptual entity, the target. Source and target are in the same frame and are linked by a pragmatic function, so that the target is mentally activated. (Adapted from Barcelona 2011.) Some comments on this definition are necessary. The first point to note is that, as noted above, metonymy does not have to be referential. Some exam- ples of non-referential metonymies will be provided in later sections. Mapping refers to the fact that the source domain is connected to the tar- get domain by imposing a perspective on it. In the sentence Picasso is not easy to appreciate, PICASSO’S ARTISTIC WORK is a metonymic target and its activation is carried out from the source Picasso, in his role as ARTIST, with the result that the hearer/reader is invited to conceptualize this artistic work primarily as the outcome of Picasso’s artistic genius—as an extension of his personality—, other aspects of this work being backgrounded (see Barcelona 2003a). The mapping in metonymy is normally asymmetrical, i.e. not a symmetric systematic matching of counterparts, as in metaphor (Barcelona 2002a, 2003a). Metonymy operates within a frame (Fillmore 1985), i.e. within a “func- tional domain” (Barcelona 2002a). The term frame is equivalent to Lakoff’s term ICM [Idealized Cognitive Model] (Lakoff 1987). In my view, the cog- nitive domain within which metonymy operates according to most standard cognitive linguistic definitions of metonymy (e.g. Kövecses & Radden 1998) is necessarily a “functional cognitive domain” (i.e. a frame or ICM), not a taxonomic cognitive domain. This allows us to understand why John is a lion is metaphorical and not metonymic, even though people and animals are in the same taxonomic cognitive domain, whereas in The White House is ready to act the subject NP is metonymic, since people (the US President and/or his government) and buildings, which are taxonomically distinct do- mains, are in the same frame or functional domain (the LOCATION frame). As regards the term pragmatic function, a fundamental property of me- tonymy is the fact that the source maps onto and activates the target by vir- tue of the experiential (hence pragmatic) link between the roles each of them performs in the same “functional domain”. This is why Fauconnier (1997: 11) regards metonymy as a “pragmatic function mapping”. A “pragmatic function” is a strong, privileged built-in connection between roles in a frame or ICM (CAUSE-EFFECT, AUTHOR-WORK, AGENT-ACTION, etc.). The term mental activation means that the source is a reference point providing mental access to the target (Langacker 1993, Kövecses & Radden 1998). The source is a privileged vehicle to mentally access the target precisely by virtue of the pragmatic function linking them. The view of metonymy as a privileged 15 connection and as mental activation is related to the view of metonymy as a shortcut (Alm-Arvius 2003). Since the goal of this article is to survey recent research on the ubiquity of metonymy, I have striven to ensure that at least most of the metonymies discussed in that research will be uncontroversially recognized as such by most linguists. The above general, “schematic” notion of metonymy covers a set of increasingly more restricted definitions proposed in Barcelona 2003a (“purely schematic”, “typical” and “prototypical” metonymies). There is no space here to discuss each of those definitions. Suffice it to say that, while purely schematic metonymies are “peripheral”, “marginal”, thus “controver- sial” metonymies close to non-metonymic (“literal”) conceptualization, typi- cal and prototypical metonymies are clear, undoubted metonymies. Only typical and prototypical metonymies have been considered for this research. Examples of the three general types: (1) This book weighs two kilograms. (2) This book is highly instructive. (3) That Mercedes is a bastard! He should be fined for dangerous driving. (4) She’s just a pretty face. (Lakoff & Johnson 1980: 37) (5) He walked with drooping shoulders. He had lost his wife. The metonymic status of the italicized parts of examples (3), (4) and (5) is uncontroversial. Example (3) is a referential metonymy with an individual entity (the driver of the Mercedes) as target. The metonymies in examples (4) and (5) are not referential. In (4), a body part activates the whole person, so that a pretty face stands for a pretty woman. The target in (5) is not an individ- ual entity, but a relation (an emotional state): What is conventionally believed to be a possible behavioral effect of sadness (walking with drooping shoul- ders) activates its cause (the emotion itself), so that an automatic metonymy- based inference is that the person exhibiting this bodily behavior was sad. The metonymies in examples (1) and (2) are more controversial. In (1), the whole domain BOOK can be argued to be mapped onto its sub-domain PHYSICAL OBJECT, which is thus mentally activated. In (2), the whole do- main BOOK is mapped onto its sub-domain SEMANTIC CONTENT, which is thus mentally activated. Examples (1) and (2) are then “peripheral” or “pure- ly schematic” instances of metonymy, whose target is a primary or near- primary domain—in Langacker’s (1987: 165) sense of “primary domain”— included in the source domain; in other words, the target is not easy to iso- late conceptually from the source. In (3), (4) and (5), the target can be more easily isolated from the source, so that a semantic shift in the use of the met- onymic expressions is noticed more readily. Purely schematic metonymies are contextual semantic values often arising in the “literal” use of expres- sions, a fact which points to the artificiality of a strict, rigid literal-figurative distinction. 16 The status of what I call purely schematic metonymies is only one of the problems affecting the cognitive linguistic notion of metonymy. Apart from the perennial problem of the distinction from metaphor, there are a few other unresolved issues which cannot be treated here (see Barcelona 2002a, 2003a, Benczes, Barcelona & Ruiz de Mendoza 2011). An issue partly overlapping with that of purely schematic metonymies is the metonymic status of active zones (AZ), since sentences (1) and (2) above are actually instances of zone activation. (On active zones and metonymy, see Langacker 2009, Paradis 2004, Geeraerts & Peirsman 2011, Barcelona 2011.) Only those AZ phe- nomena which are clearly metonymic as well have been considered in this paper. Examples (1) and (2) are peripheral metonymies. An example of a clearly metonymic AZ is (6), which constitutes a prototypical metonymy: (6) The Times criticized the minister. [COMPANY (NEWSPAPER) FOR EM- PLOYEE (JOURNALIST)]. Compare this with instances like Langacker’s (1999: 62) example, repro- duced as (7) below, which is doubtfully metonymic: (7) Your dog bit my cat. [ANIMAL (DOG) FOR ANIMAL’S BODY PART (DOG’S SET OF TEETH)? and ANIMAL (CAT) FOR UNSPECIFIED PART OF ANIMAL’S (i.e. CAT’S) BODY?] 3 Metonymy in conceptualization: cognitive models, meta- phor, blending There are a number of generally “invisible” conceptual operations or con- ceptual structures (i.e. operations/structures not directly coded by a particular linguistic form) that underlie online linguistic processing and that motivate linguistic structure at all levels. The frequent involvement of metonymy in them is evidence both of the conceptual nature of metonymy and of the fact that metonymy is not confined to lexical meaning. In the rest of this section I briefly report on the role of metonymy in a conceptual structure (cognitive models) and in two closely related types of conceptual operations (metaphor and blending). 3.1 Metonymy in cognitive models Lakoff (1987: Chapter 5) was one of the first linguists to point out that a great many category models (i.e. ICMs, frames) are organized around a me- tonymy-based prototype. These models he called metonymic models. They pervade our cognition and our language. In a metonymic model, the proto- 17 type is a sub-category acting as a metonymic reference point (Langacker 1993) for the whole category. The metonymies organizing the models are normally invisible metony- mies, since in most cases they do not directly motivate specific linguistic forms; rather they motivate the whole cognitive model. Metonymic models, like all other cognitive models, govern aspects of our reasoning and our comprehension of discourse (especially in pragmatic inferencing). As examples of metonymic models we can briefly mention two well- known “social stereotypes” (Lakoff 1987: Chapter 5):  The MOTHER stereotype, where the “housewife-mother” (i.e. the mother that does not work outside the home) stands for the whole MOTHER category. Note that there is no conventional lexeme naming this model. The model underlies the type of reasoning represented by the following sentence: Jenny is a good mother even though she  has a demanding job. The BACHELOR stereotype (characterized by dating a lot, frequenting singles bars, being only interested in sexual conquest, etc.), which metonymically “overshadows” the whole BACHELOR category. This model motivates examples like Marв’s husband is a real bachelor. This latter sentence is not a contradiction in the metonymic model. It would be one in the “standard”, “rigid” BACHELOR model, whose prototype is an adult, unmarried male who has, however, reached a marriageable age. There have been a handful of other studies of metonymic models by cog- nitive linguists and by other scholars. A further example is Feyaerts’ (1999) study on the metonymic models for STUPIDITY. For psycholinguistic studies on the role of metonymy in cognitive models, see Gibbs (1994, 1999, 2007). 3.2 Metonymy as a motivation for metaphor Metonymy has been claimed by Barcelona (2000a) and Radden (2000), as well as other linguists, to motivate a great many metaphors, i.e. to connect them to their experiential basis. According to Barcelona (2000a), two major types of metonymic motiva- tion of metaphor can be discerned: A. Abstraction of a common conceptual structure between metaphoric source and target. This abstraction occurs when both the metaphoric target and the metaphoric source are conceptualized metonymically from the same “subdomain”. Con- sider examples (8) and (9): 18 (8) That’s a loud color. The metaphor licensing the combination of loud with color can be called DEVIANT COLORS ARE DEVIANT SOUNDS. This metaphor is possible thanks to the abstraction of a common “subdomain”, namely the effect on perceivers of both deviant colors and of deviant, “loud” sounds. This effect is that of “irresistibly attracting the perceiver’s attention”. The abstraction of this commonality is due to the conceptually prior metonymic understanding of both DEVIANT COLORS and DEVIANT SOUNDS (as metonymic targets) from their typical effect, IRRESISTIBLY ATTRACTING THE PERCEIVER’S ATTENTION (as metonymic source). (9) acorn cup The metaphoric understanding of the lower part of this fruit as a CUP is mo- tivated by the conceptually prior metonymic understanding both of CUPS and the relevant part of ACORNS in terms of their SHAPE. B. Generalization or decontextualization of a metonymy. The well-known metaphors MORE IS UP (as manifested by countless exam- ples such as the high cost of living / skyrocketing prices / a low level of intel- ligence, etc.) is in fact the result of the decontextualization of the metonymy LEVEL OF VERTICALITY FOR QUANTITY. This metonymy is in turn due to the frequent experiential association of LEVEL OF VERTICALITY with QUANTITY in POURING or HEAPING events. Once LEVEL OF VERTICALITY is used to de- note QUANTITY in contexts where no real verticality is involved the metony- my has been decontextualized and become a metaphor. Take a low level of intelligence: intelligence is not a spatial object that can be measured in terms of spatial verticality. 3.3 Metonymy in blending See Turner & Fauconnier 1995, 2000, Fauconnier & Turner 2002. These authors argue for the existence of a number of “Optimality principles on integration networks”, one of which is the so-called “Metonymy projection constraint”. This constraint reads like this: “When an element is projected from an input to the blend and a second element from that input is projected because of its metonymic link to the first, shorten the metonymic distance between them in the blend”. An example is the conventional representation of Death as the Grim Reaper. This symbol arises by blending many input spaces. In the blend, a “long distance” metonymy connects death to the cowl of the priest assisting a dying person. 19 4 Metonymy under the lexicon: Phonemic categories and morphemic meaning 4.1 Metonymy in phonology: Motivation of internal links in phonemic categories Metonymy in phonology is briefly discussed in Barcelona (2002b) and in Radden (2005). Taylor (1995: 222–239), following Nathan (1986), claims that, in terms of Jones’s phonological theory (Jones 1969), it is possible to describe a pho- neme as a category with a network of allophones organized around a proto- typical allophone. As an example, Taylor claims that the link between the prototypical allophone of English /t/, namely [th] (the voiceless aspirated alveolar plosive) and the glottal stop [?] (which is a possible allophone of /t/ in certain phonetic contexts) is the glottalized t-allophone [?t]. A possible metonymic motivation for this connection is proposed by Bar- celona (2002b). If phonemes are conceptual categories, then they are subject to conceptual metonymy, which typically operates within categories (Lakoff 1987: Ch. 5). The glottal stop allophone [?] activates the glottal t-allophone [?t]. The alveolar closure element in the glottal t-allophone [?t] mentally [th]). Thus the chain of conceptual connections [?]  [?t]  [th] can be ar- activates the central allophone (the voiceless aspirated alveolar plosive t gued to be metonymic. This chain facilitates (together with co-textual fac- tors) the recognition of [?] as a peripheral member of /t/ by language users. According to the phonologist Geoffrey Nathan (1996), conceptual con- nections “akin to” metaphor and metonymy may play an important role in the structure of phonological categories. 4.2 Metonymy in morphemic meaning: Motivation of certain deriva- tional morphemes For metonymy in morphemic meaning, see Barcelona (2005a, 2009a, in preparation), Radden (2005), and Palmer et al (2009) on Tagalog ka-. An example (analyzed in Barcelona 2005a: 320–21) is (10), from Eugene O’Neill’s Long Daв’s Journeв into Night. (10) TYRONE. You’re a fine armful now, Mary, with those twenty pounds you’ve gained. The derivational morpheme {ful} in armful derives nouns from nouns. It originates in the adjective full. The meaning of the morpheme is ‘the quanti- ty of X that fills or would fill Y’ (definition adapted from Webster’s and the 20 OED). Other instances of derivation by means of the same morpheme are bottleful, boxful, canful, worldful, churchful etc. The standard sense of the morpheme seems to be based on the metonymy DEGREE OF FILLING OF A CONTAINER FOR QUANTITY OF CONTENT FILLING IT (a manifestation of VERTICALITY FOR QUANTITY) (Barcelona 2000a; Radden 2000). The domain of quantity is activated by the degree to which the con- tainer has been filled by the content, within the FILLING frame. The container can be metaphorical (a person’s arms in armful) or non- metaphorical (as in bottleful). The container is specified in the derived noun by the lexical morpheme ({arm}, {bottle}). The metonymy can be justified on independent grounds, as it can moti- vate other conventional linguistic expressions where the degree of filling of a container is a metonymic source for a certain quantity of content: Give me half a glass of beer. This morpheme is different (or has a different sense) in playful, where it derives adjectives and it means ‘full of’, ‘having’, ‘characterized by’. This morphemic meaning is metaphorically, not metonymically motivated: An abstract notion, typically a process (PLAY, JOY, etc.) is understood as a phys- ical “content”, and a person, a behavior (a hug), etc. is understood as a meta- phorical container: A beautiful lady / He gave her a playful hug. 5 Lexical metonymies are not just lexical Numerous examples of lexical metonymies are offered in the section on metonymy in grammar, below. Lexical metonymies are, after all, extremely frequent. In this paper, I am not primarily interested in only the lexical- semantic effects of lexical metonymies; I am more interested here in the grammatical side of lexical metonymies (especially as regards the morpho- syntactic categorization of the respective lexemes; section 6 below) and in their discourse function (especially as regards their role in discourse- pragmatic inferencing and, in general, in the creation of discourse coher- ence). The goal of the present section is simply to offer a few brief comments on the linguistic behavior of lexical metonymies, to show that they are often inseparable from metonymies operating at higher grammatical levels or even at discourse level. This fact adds to the fact that metonymies affecting main- ly or exclusively lexical items are probably less frequent than text-level purely inferential metonymies (Barcelona 2005a, b) ranging over whole sen- tences or over whole propositions (see section 7) and to the fact that meton- ymy also often operates at sub-lexical levels (morphology and probably also phonology), as claimed in section 4. Taken together, these facts should con- 21 tribute to undermining the widespread view of metonymy as an exclusively or almost exclusively lexical phenomenon. If we are asked to point out examples of lexical metonymies, we will normally produce such sentences as (11). (11) The buses (i.e. the bus drivers) are on strike. This metonymy is an instance of the metonymic pattern CONTROLLED FOR CONTROLLER; see Radden & Kövecses 1999. This seems to be evidence that the unconscious model or prototype of the “lexical metonymy category” is represented by nominal referential metonymies (i.e. metonymies operating over nouns). But this is not necessarily the case. Firstly, lexical metonymies are not necessarily restricted to nouns: (12) I am parked over there. (Radden & Kövecses 1999) The metonymy operates over I, which is a personal pronoun. We can also find metonymies operating over other lexical classes. In section 6 below we find instances of metonymies operating over an adjective and converting it into a noun (interstate), over a cardinal numeral determiner and converting it into a cardinal numeral pronoun (twenty, sixty-five), or over a verb and caus- ing its subcategorial conversion (reduce), to name but a few examples of those gathered in that section. Secondly, when nominal metonymies are referential, they necessarily op- erate over phrases rather than just over their nominal head. In (11) “the bus drivers” in question are the intended referents, not of buses, but of the whole NP The buses, even though the referential value of the latter crucially de- pends on the metonymized lexical meaning of its head. That is, these meton- ymies are not just lexical; they are also phrasal. Thirdly, whether nominal or not, and whether referential or not, metony- mies normally affect grammatical structure beyond lexemes and discourse comprehension—further evidence that they are not just lexical. In (12) and in all the cases in section 6 mentioned above (interstate, twen- ty, sixty-five, reduce), the shift in lexical sense or in reference effected by metonymy brings about a change in the syntax of both the respective lexeme and the syntactic constructions where it enters. The choice of I instead of My car in (12) causes the selection of a first person singular form of the verb (I am parked rather than My car is parked). The metonymy-driven conversion of reduce from a causative transitive into an intransitive verb modifies the argument structure of the clauses governed by this verb (no patient-direct object slot is available now). In other words, the lexical metonymies (nomi- nal or otherwise) bringing about some type of lexical recategorization also affect phrasal or clausal syntax; hence they are not purely lexical. 22 As for the discourse function of lexical metonymies, referential metony- mies are often instrumental in achieving referential coherence (i.e. they help hearers/readers to identify the intended referents throughout an extended piece of discourse; see e.g. Barcelona 2005a and 2007 for examples). These metonymies point to the intended referent on the basis of inferencing (hence discourse-pragmatic inferencing is often involved; see Warren 1999: 123). Lexical metonymies are at the same time instrumental in achieving relation- al coherence (i.e. the conceptual connections among the various sentences in a discourse: cause-effect, exemplification, condition-result, elaboration, jus- tification, etc.);2 lexical metonymies often serve this purpose by chaining to other high-level metonymies (Barcelona 2005a, b, 2007). An example (bor- rowed from Radden 2000 and analyzed in Barcelona 2005a) is (13): (13) A: How much gas did you buy? B: I filled ’er up. The inferences by A that B filled the gas tank and that (s)he filled it with the maximum amount that the tank could hold are guided, among other factors, by the lexical metonymies CAR FOR GAS TANK (WHOLE FOR ACTIVE ZONE PART)3 and UP FOR MOST (an instance of VERTICALITY FOR QUANTITY). The first of these metonymies is involved in referential coherence and also in relational coherence, since the correct identification of intended referents is often a prerequisite for achieving relational coherence, as in this case (estab- lishing the gas tank as the container that was filled up is a prerequisite for making the response consistent with the reply and A’s normal expectations in terms of cooperative maxims). But these metonymies, of course, are not just lexical; they are also discourse-level metonymies. For the notions of referential and relational coherence see Dirven & Verspoor (1998: Ch. 8). In sum, lexical metonymies are often not easy to distinguish from phrasal metonymies, from other grammatical metonymies (on the other hand, the dividing line between the lexicon and other grammatical units is often fuzzy) and from text-level metonymies. This fact, together with the fact that meton- ymy often operates under and above the lexicon, supports the above claim that metonymy is not just, or even fundamentally, a lexical phenomenon. Metonymy is above all a conceptual connecting device between elements in our experience, which are often, but not necessarily, coded lexically in language, and it is often not confined to one particular grammatical level or even to grammar. It is a ‘natural inference schema’ (Panther & Thornburg 2 These connections, when not explicitly marked linguistically, are often inferred (implica- tures) with the help of metonymies (lexical or otherwise); see section 7. 3 The car is referred to metaphorically by means of (h)er. 23 2003a) operating at many different linguistic analytical levels (Barcelona 2005a, 2005b, 2007, 2009a, in preparation). 6 Metonymy in grammar (lexemes and above) 6.1 A few general comments Langacker (1999: 67) says that though “usually regarded as a semantic phe- nomenon, metonymy turns out to be central and essential to grammar”, and that grammar is “a rich source for the investigation of metonymy. At the same time, a recognition of its prevalence and centrality is critical not just for describing grammar but for a realistic assessment of its basic nature”. I could not agree more. In a large part of my more recent work (Barcelona 2002b, 2005a, b, 2009a, in preparation) I have argued and provided indirect evidence for the claim that metonymy is a multi-level phenomenon, often simultaneously present at more than one grammatical level in a given sentence. However, despite metonymy’s pervasive role in grammar, Brdar argues that it is involved only in the nitty-gritty details of grammatical structure. (2007b: 69). According to him, metaphor has a more wide-ranging impact on grammar. Important recent surveys of the interaction between metonymy and grammar are Brdar (2007b), Langacker (2009), Panther, Thornburg & Bar- celona, eds (2009), Ruiz de Mendoza & Pérez Hernández (2001) and Ruiz de Mendoza & Otal (2002). The rest of this section is devoted to reporting on recent research (includ- ing my own) on the various areas of interaction between metonymy and grammar mentioned in section 1 (recategorization, etc.) and to providing a detailed example of my own research in one of these areas. Due to limita- tions of space, the report presents only a representative subset of the rich work in this field. 6.2 Grammatical recategorization (transient or permanent) In this sub-section we examine some instances of shifts of the grammatical class of a lexical item (but also of a phrasal or a syntactic construction) mo- tivated, together with other factors, by metonymy. Some of these shifts do not always become permanent, i.e. do not necessarily result in a stable recat- egorization of the lexeme or construction involved (6.2.1). But others nor- mally do, so that the same basic form now realizes a different lexical class: these are the instances of conversion, affixal derivation and grammaticaliza- tion illustrated in 6.2.2. 24 6.2.1 (Relatively) transient grammatical recategorization A. Statives-as-dynamics For statives-as-dynamics, see Panther & Thornburg 2000, Ruiz de Mendoza & Pérez Hernández 2001, Ruiz de Mendoza & Otal 2002. An example is (14): (14) He asked her to be his wife. According to Panther & Thornburg (2000), who studied this and similar instances of such recategorization, the above sentence could be paraphrased as ‘He asked her to act in such a way so as to become his wife’. They claim that the shift is motivated by the metonymy EFFECT FOR CAUSE: “Be some- one’s wife” is a “resultant state”, which is controllable (Ruiz de Mendoza & Pérez Hernández 2001) and a metonymic source for its implicit causal ac- tion. When a state cannot be the result of an implicit causal action it cannot be a metonymic source for the latter: *He asked her to be tall. Nor can a resultant state be a metonymic source for its implicit causal action if it is not controllable. Cf. *Fall asleep (“falling asleep” is not a controllable state), as opposed to Don’t fall asleep (“not falling asleep” is a controllable state). B. Transient (occasionally permanent) recategorization of English proper names as common nouns See Barcelona 2003c, 2004, Brdar 2007a, b, Soloshenko 2005, inter alia: (15) John has five authentic Picassos (AUTHOR FOR WORK). The above metonymy and similar types (such as PRODUCER FOR PRODUCT, as in I bought a Mercedes) account for many such recategorizations. Let us illustrate two special cases on which I have done some careful re- search. The discussion of both of them is actually much more complex; I can only present a highly simplified account here: B1. Paragons (Barcelona 2003c, d, 2004, Brdar 2007a, b, Pang 2006) (16) That graduate student is an Aristotle. There aren’t any real Aristotles today. A real Aristotle would be more systematic. 25 I have claimed (Barcelona 2004) that the transient recategorization of Aristo- tle as a common noun depends on a conceptual process with the following elements: a) Stereotypical understanding of the individual Aristotle in terms of his salient characteristic property, i.e. being a great philosopher, this stereotypical understanding being due to the metonymy CHARAC- TERISTIC PROPERTY (‘being a great philosopher’) FOR ENTITY. b) Creation of a class of distinct individuals connected by sharing the same characteristic property, the ideal member of which (the para- gon) is Aristotle. c) Metonymic activation of the mental class by its most outstanding member (STEREOTYPICAL MEMBER FOR CATEGORY). This second metonymy is directly responsible for the transient grammatical re-categorization of this name as a common noun (which allows the name to take a number morpheme and to be preceded by determiners and to be ac- companied by restrictive modifiers). The recategorization is sometimes permanent and enters the dictionary as an ordinary common noun (and is even spelled with a lower-case initial let- ter): Caco (mythological paragon of thieves) becomes the Spanish common noun caco (colloquially, ‘thief’). Metonymy is also involved in the emer- gence of the Spanish common noun galeno (‘doctor’), from Galen (Galenus in Latin), a famous medical doctor of the Roman period, whose theories dominated Western medical science for over a millennium. B2. Partitive restrictive modification (Barcelona 2003c, Barcelona 2003d, Barcelona 2009b, Brdar 2007a, b) (17) The young Joyce already showed signs of the genius that was to be fulfilled in Ulysses. The older Joyce was wiser than the young Joyce. (Examples borrowed from Greenbaum & Quirk (1990: 88).) We find in these cases an instance of the partition of an otherwise unitary entity (James Joyce) into a fictitious class of individual “entities”, each of them represent- ing a semantic relation involving Joyce (“Joyce when young”, “Joyce as an older man”, etc.); however, in the two sentences of (17) the speaker always has the same one real-world referent in mind, James Joyce. This is an in- stance of what Fauconnier & Turner (2002) call “decompression”. Cf. the difference with paragon names: the latter can be used to refer to individuals different from the paragon (i.e. other than Aristotle). In my above-mentioned articles (especially in Barcelona 2009b), I have suggested that the metonymy ENTITY (JOYCE) FOR ACTIVE ZONE RELATION 26 (JOYCE WHEN YOUNG) motivates this construction, together with the analogy of this construction to names with non-partitive restrictive modification such as the fondly remembered John F. Kennedy. This metonymy is a purely con- ceptual, pre-linguistic mapping. In Barcelona (2009b) I also answer some possible objections to this account. Few of the instances of recategorization treated in this sub-section have entered the dictionary. Most of these cases are still viewed as special uses of names (occasionally behaving as common nouns), not as firmly established types of common nouns. 6.2.2 (Relatively) permanent grammatical recategorization A. Conversion A1. Mass noun-count noun conversion (Brdar 2007b, Ruiz de Mendoza & Pérez Hernández 2001, Ruiz de Mendoza & Otal 2002) CountMass: MATERIAL FOR OBJECT MADE OF THAT MATERIAL (Brdar 2007b; see also Kövecses & Radden 1998): (18) We did not always eat turkey for Christmas dinner. MassCount: OBJECT FOR MATERIAL CONSTITUTING THE OBJECT (ibid): (19) To have won one gold medal and two silvers in those Games was his- toric. (Examples taken from the OED and cited by Brdar 2007b: 79.) A2. Noun-verb conversion (Dirven 1999, Ruiz de Mendoza & Pérez Her- nández 2001, Ruiz de Mendoza & Otal 2002) Examples taken from Dirven (1999: 275–287), with the metonymies sug- gested by him: (20) He was angling. (INSTRUMENT FOR ACTION) (21) The plane was forced to land in Cairo. (GOAL FOR MOTION) (22) Mary nursed the sick soldiers. (CLASS MEMBERSHIP FOR DESCRIPTION) In my view, (22) can be accounted for simply as an example of AGENT FOR ACTION. In Barcelona (2009a, in preparation), I cite the following example: (23) I parked my car carefully. 27 The conversion is due to the metonymy GOAL FOR MOTION, on the basis of one of the senses of the noun park (Webster’s: ‘a space set aside for leaving vehicles temporarily’; see also OED 5b). A2. Adjective-noun conversion Interstate (adjective) freeway/highway  (an/the) Interstate (noun); both lexemes originate in American English. The adjective means “between (U.S.) States”, according to Webster’s, and can function almost exclusively as a syntactic modifier never taking the plu- ral morpheme. The noun means “interstate freeway/highway”, and it behaves like a typical noun, as can be seen by these examples: (24) All interstates are necessary. The federal government is building an expensive new interstate. The interstate’s length is 2000 miles. The new interstates increased the mobility of the American people. The conceptual metonymy: SALIENT POLITICAL-GEOGRAPHICAL PROPERTY OF A HIGHWAY (LINKING TWO STATES) FOR THE HIGHWAY. This metonymy is a special case of the higher-level metonymy SALIENT PROPERTY OF A CATEGORY FOR THE CATEGORY. A3. Cardinal numeral determiner-cardinal numeral pronoun conversion (Barcelona 2009a, in preparation) (25) TYRONE. But thank God, I’ve kept my appetite and I’ve the digestion of a young man of twenty, if I am sixty-five. (Example taken from Eu- gene O’Neill, Long Daв’s Journeв into Night, Act 1, Scene 1.) The conversion is at least in part due to metonymy-motivated ellipsis: SALI- ENT PART OF FORM FOR WHOLE FORM. The salient parts of the form of the two age-measuring NPs in the example (twenty years of age; sixty-five years old) are the numerals, which are more salient than the other elements in the NP because they are the first element in the phrase and because they are, semantically, more informative. A4. Other types of conversion In Barcelona (2009a, in preparation), I have identified several instances of metonymy-motivated conversions of a lexical subcategory into another lexi- cal subcategory of the same lexical category. One of these cases is the shift from the interrogative adverb why to the absolute relative adverb why. The 28 metonymy involved in the shift is INTERROGATION ABOUT THE REASON FOR A STATE OF AFFAIRS FOR THE REASON FOR THAT STATE OF AFFAIRS. B. Affixal derivation B1. Deverbal nominalization (Panther & Thornburg 2002, Szawerna 2005, Palmer et al. 2009) An example is the derivation of the noun cliffhanger with the meaning ‘sus- penseful event’ (Panther & Thornburg 2002). The metonymies involved are the following:  (SCHEMATIC) HUMAN PARTICIPANT (EXPERIENCER) FOR (SCHE- MATIC) EVENT (in which the human participant is crucially in- volved). This metonymy operates on the meaning of the {er} mor-  pheme as designating an unspecified event. The specific event (hanging from a cliff) is designated by the lexical base and this event is further elaborated by CAUSE (the activity of cliffhanging) FOR EFFECT (suspense). B2. Noun-noun derivation See the above comment in section 4.2. on the special sense of the derivation- al morpheme in (an) armful, (a) bottleful, etc. (Barcelona 2009a, in prepara- tion). C. Other types of recategorization An example is the metonymy-based recategorization of an NP as a phrasal name, as in the use of the noun phrase The Continental Divide as a conven- tional name for the Rocky Mountains (Barcelona 2009a, in preparation). The metonymy motivating in part the recategorization is SALIENT PROP- ERTY OF AN ENTITY (function as Continental Basin divide) FOR THE ENTITY (the Rocky Mountains). This metonymy operates within the NORTH AMERI- CAN GEOGRAPHY frame. D. Metonymy-motivated grammaticalization Metonymy, in combination with metaphor, is often a prime factor in gram- maticalization (Traugott & Dasher 2002, Hopper & Traugott 1993). An oft- cited example is the metonymic basis of the implicatures leading to the grammaticalization of the expression be going to as a marker of futurity and/or intentionality (Hopper & Traugott 1993, Heine, Claudi & Hünnemey- er 1991). 29 Recent examples from metonymy researchers: D1. Body-part noun  Grammatical word Hilpert (2007) claims that in some cases, metonymy is part of the chain of extensions leading to the grammatical meanings (which bring about the con- version of these nouns into prepositions, postpositions, or adverbs) devel- oped in many languages by nouns whose basic meanings are ‘back’, ‘belly’, ‘buttocks’, ‘face’, ‘forehead’, ‘hand’. For example, in a number of languages ‘back’ develops the grammatical meanings ‘behind’ (of an object), and (temporal) ‘after’ thanks to the chaining of the metaphor OBJECTS ARE HU- MANS, the metonymy PART FOR REGION and the metaphor TIME IS SPACE. D2. Inert object noun  Multal quantifier noun Barcelona (2009a, in preparation) claims that a long chain of metonymies interacting with one metaphor motivates in part the development of the mul- tal quantifier meaning of lot (originally designating an object used in decid- ing a matter by chance, usually a piece of wood; see OED, sense 1a of lot) in the quasi-determiner expression a lot of and the quasi-pronoun a lot. 6.3 Compounding Metonymy (often in association with metaphor) determines the choice of one or both bases in a large number of compounds (Radden 2005, Barcelona 2008, Benczes 2006, Kosecki 2005a). A particularly important example is constituted by bahuvrihi compounds, which are a type of exocentric compounds (Barcelona 2008). The ability of these nominal compounds to denote a category of entities (people, people, plants, animals, objects, etc.) is due to the overriding metonymy CHARAC- TERISTIC PROPERTY FOR CATEGORY. For example, the noun humpback de- notes a category of people by mentioning the characteristic property exhibit- ed by that category. However, the semantic structure of these compounds is often fairly com- plex, since, according to my analysis, the characteristic property mapped by the above-mentioned overriding metonymy is seldom conceptualized literal- ly (as in humpback), being in most cases conceptualized metonymically, i.e. by means of other metonymies (as in hardtop, wetback) or metaphtonymical- ly, i.e. by means of the interaction of other metonymies and one or more conceptual metaphors (as in highbrow, featherweight, or blockhead). 30 6.4 Phrasal grammar: Generic NPs Radden (2005, 2009) claims that all of the four types of generic NPs that he discusses are motivated in part by metonymy and conceptual blending (see also Radden & Dirven (2007: 107–111)). Let us consider a few examples of each (borrowed from Radden 2009). Radden’s treatment is very detailed and insightful. For lack of space I can only mention next to each example the type of generic NP, with its grammatical form in brackets) and my reading of the motivating metonymy suggested for each of them by Radden (2009): (26) A lion has a bushy tail. Type: Representative generic (indefinite singular) Metonymy: (ARBITRARY) INSTANCE FOR TYPE (27) Hedgehogs are shy creatures. Type: Proportional generic (indefinite plural) Metonymy: (PROPORTIONAL SET OF) INSTANCES FOR TYPE (28) The tiger hunts by night. Type: Kind generic (definite singular) Metonymy: (PROTOTYPICAL) INSTANCE FOR TYPE (29) The Italians love pasta. Type: Delimited generic (definite plural) Metonymy: (DELIMITED SUBTYPE OF) INSTANCES FOR TYPE There is another type of delimited generics with adjectives applicable to people as NP heads (as in The young are taking over now) that are motivated by the metonymy PROPERTY FOR THE ENTITY THAT HAS THE PROPERTY. 6.5 Metonymy in the tense-aspect-mood system of English verbs In certain cases, metonymy allows English speakers to use a modal verb phrase introduced by can, which literally profiles a potential future situation, to designate a situation occurring at the time of utterance, a present situation. This typically happens with verb phrases headed by stative verbs of percep- tion or sensation (e.g. feel), which cannot normally be used in progressive verb phrases, to indicate that the situation described occurs at the time of utterance. Panther & Thornburg (1999) claim that this shift is motivated by the metonymy POTENTIALITY FOR ACTUALITY: (30) I could hear his sneering laughter as her arms carried me off through the fire of oblivion. I can hear it yet. (Approximate paraphrases: “I heard his sneering laughter... (at that moment) / I still hear it (at this moment)”). (31) I can taste the vanilla (Approximate paraphrase: “I taste vanilla (at this moment)”). 31 Examples borrowed from Panther & Thornburg (1999: 340); example (30) was taken by Panther and Thornburg from the Lancaster-Oslo-Bergen cor- pus. Brdar (2007b: Ch 7) claims that the widespread use of the present tense for habitual and timeless situations is ultimately motivated by metonymy. One simple example: (32) Mary speaks Spanish. The metonymy behind this extension is called by Radden & Kövecses (1999) PRESENT FOR HABITUAL. 6.6 Clausal grammar According to Ziegeler (2007), a metonymic account seems in general to be a better alternative than simple “coercion” analyses. She gives convincing arguments for the claim that “nominal coercion”, as proposed by Michaelis (2004), and “complement coercion” and “subject coercion”, as discussed by Pustejovsky (1995) can directly be accounted for in terms of metonymy. Nominal coercion is claimed by Michaelis to occur in examples like She had a beer or You have apple on your shirt. (“Coercion” means that it is only the grammatical construction—the syntax—that determines the reinterpretation of beer as a count noun or of apple as noncount.) Ziegeler argues very pow- erfully that these shifts can more simply be explained in terms of metonymy (CONTAINED FOR CONTAINER for the beer example and OBJECT FOR MATE- RIAL CONSTITUTING THE OBJECT for the apple example). As for Pustejovsky’s claim, Ziegeler suggests a metonymic account which is, in fact, an account in terms of active zone metonymies. Langacker (1999, 2009), Brdar (2007b), Ruiz de Mendoza & Pérez Hernández (2001) and Ruiz de Mendoza & Otal (2002)—the latter two under the label “metonymy in the predication”—have discussed examples similar to Pustejovsky’s: (33) George began the book. He wanted to read the whole of it. (34) A: What could we buy for Mary’s birthday? B: A book would be a good idea. Not all of these authors use the term active zone, a term due to Langacker (1987), but all the metonymies they suggest are actually active zone meton- ymies: We could respectively call them BOOK FOR RELATION [X READ BOOK] in (33) and BOOK FOR RELATION [X BUY BOOK] in (34). Many other aspects of clausal grammar have been shown to be motivated by metonymy. We can only present two illustrative examples here: 32 A. Metonymy and valency extension and reduction in relation to transitivity Brdar (2007b), Ruiz de Mendoza & Pérez Hernández (2001) and Ruiz de Mendoza & Otal (2002)—again under the label “metonymy in the predica- tion”—, Ruiz de Mendoza & Mairal (2007) and Barcelona (2009a, in prepa- ration) have discussed this issue. A simple example which I have discussed elsewhere (Barcelona, in prepa- ration, 2009a) is the valency reduction and shift in transitivity status (this shift also being an instance of subcategorial conversion) of a verb like re-  Causative transitive meaning of the verb (Webster’s: ‘to lessen in duce; the reduction and shift take place in two successive steps: any way, as in size, weight, amount, value, price, etc; to diminish’)  (ii) General intransitive meaning of the verb (Webster’s: ‘becom- ing reduced in general’). This shift is motivated by the metonymy ACTION [CAUSING X TO BECOME REDUCED IN GENERAL] FOR RESULT  (ii) General intransitive meaning of verb (‘becoming reduced in gen- [X BECOMING REDUCED IN GENERAL]. eral’)  (iii) specific intransitive sense (Webster’s: ‘lose weight, as in by dieting’, that is ‘to become reduced in weight’). This shift is motivated by the metonymy CATEGORY [BECOMING REDUCED IN GENERAL] FOR MEMBER [BECOMING REDUCED IN WEIGHT]. Meaning (i) is older than (ii), and (ii) is older than (iii), according to the OED. As can be seen, the shifts (i)–(iii) are motivated by a metonymic chain (Barcelona 2005a; Hilpert 2007). B. Modality Ruiz de Mendoza and his co-workers (among others) have discussed this issue. An example is the metonymy OBLIGATION FOR DESIRE TO CARRY OUT THE DESIRED ACTION as in the volitional use of must: I must speak to you, please (Ruiz de Mendoza & Pérez Hernández 2001). 6.7 Anaphora The role of metonymy in anaphora has been discussed in particular by Lan- gacker (1999, 2009), Ruiz de Mendoza & Pérez Hernández (2001) and Ruiz de Mendoza & Otal (2002). A. Metonymy-induced pronominal anaphor Langacker (1999: Chs. 7 and 9, 2009) claims that the only real factors con- straining anaphora are conceptual rather than syntactic. He discusses in- stances in which the intended antecedent of the anaphor is a metonymic tar- 33 get of a reference point active in the mental space currently active discur- sively: (35) He speaks excellent French even though he’s never lived there. The intended antecedent of there is the metonymic target of French (i.e. France). The metonymy guiding the inference of that antecedent is a PART (LANGUAGE) FOR WHOLE (COUNTRY) metonymy. Metonymy is involved in other types of indirect anaphora (Emmott 1997), including “bridging inferences” (Clark 1977): (36) I saw a wonderful car in the street. The steering wheel was made of ivory. The NP the steering wheel is definite, due to its metonymic connection to the referent already introduced by a wonderful car. The steering wheel can only be the steering wheel of the car just mentioned; hence it is a unique referent in this context. This inference is guided by the metonymy WHOLE [CAR] FOR PART [STEERING WHEEL]. The car is a metonymic reference point for its tar- get part. 6.8 Grammatical formal metonymy Language users store in their minds idealizations, models or schematizations (Langacker 1987) of linguistic forms at all levels. These idealizations are conceptual in nature. There are normally one or more canonical/prototypical forms for a construction4, and a number of non-canonical forms (e.g. gaso- line vs. the clipped form gas). We cannot discuss here the issue of whether or not the pairing of each of the forms in the set with their meaning actually constitutes a different (yet related) construction. In any case, such a set of forms can be claimed to constitute a small frame, or a small category (i.e. a mental representation). If constructional forms constitute conceptual catego- ries, then they can be subject to a conceptual operation like metonymy, which typically operates within categories. This issue has been dealt with in close detail by Bierwiaczwonek (2005); see also Radden & Kövecses 1999, Kövecses & Radden 1998, Barcelona 2005a, b, and Barcelona, in prepara- tion, 2009a. The constructional forms that are claimed by these authors to be motivated by metonymy are mainly certain types of lexical abbreviations (in writing and speech), certain weak forms, and certain types of ellipsis. We only have space here to provide two simple examples (Barcelona in preparation, 2009a): 4 In the broad sense of the term construction in cognitive linguistics, as designating any con- ventional form-meaning pairings from morphemes upwards; see Goldberg (2006). 34  The abbreviated form gas of the noun gasoline. The non-prototypical form is licensed by the metonymy SALIENT PART OF FORM FOR WHOLE FORM. The form gas is “salient” because the spoken seg- ment g&s and the written segment <gas> are the most prominent segments of their respective full forms. They have initial position and, in the spoken canonical form, g&s bears primary stress; and semantically, both segments are readily capable of evoking the basic meaning of the lexeme, unlike li;n or <-line>.  The elliptical form of a complex clause: Matrix (main) clause ellip- sis. Let us take this well-known joke, attributed to W.C. Fields: (37) Speaker A: Do you believe in clubs for young men? Speaker B: Only when kindness fails. For a full analysis of this exchange, see Barcelona (2003b and in prepara- tion). The abstract canonical form of this complex clause is ‘Matrix part of the main clause + when clause’, i.e. I believe in clubs for young men only when kindness fails.5 The ellipsis is formally possible because a subordinate clause introduced by a subordinator (when) can act as a metonymic source for the whole complex clause: SALIENT PART OF (CONSTRUCTIONAL) FORM FOR WHOLE (CONSTRUCTIONAL) FORM. The metonymy helps Speaker A to recognize Speaker B’s utterance as an elliptical form of the complex clause. 6.9 A detailed example of a metonymy-motivated grammatical con- struction: Epistemic conditional constructions I have treated this issue in detail in Barcelona 2006 and also in Barcelona in preparation, 2009a. In epistemic conditional constructions, the speaker draws a conclusion on the basis of the premise supplied by the if-clause. We find one such construction in the following sentence: 5 In my (fairly traditional) terminology, a complex clause is a clause consisting of a main clause and a subordinate clause; the main clause normally includes the subordinate clause, its “matrix” being the main clause itself without its subordinate clause. A sentence may consist of one simple clause, one complex clause (in this case the sentence can also be called “com- plex”, as in I believe in clubs for young men only when kindness fails), or two or more inde- pendent simple or complex clauses in juxtaposition or coordination (it would then be a “com- pound” sentence, as [I believe in clubs for young men only when kindness fails] but [my belief may be utterly wrong], where two independent complex clauses stand in coordination within the same compound sentence). 35 (38) If you have ever driven west on Interstate 70 from Denver to the Con- tinental Divide, you have seen Mount Bethel. (‘Mount Bethel’ 1996) The sentence in (38) consists of a complex clause, which in turn includes a subordinate conditional if clause. We are concerned with the constructional meaning of the whole complex clause. This meaning is one of epistemic necessity, approximately paraphrased like this: ‘If I can assume it to be a true fact that you have ever driven west on Interstate 70 from Denver to the Con- tinental Divide, then I necessarily have to conclude that you have seen Mt. Bethel’. Eve Sweetser (1990: 116–117, 1996) claims that epistemic conditionals are a metaphorical extension from “normal” conditionals, the latter being called content conditionals by her and predictive conditionals by Dancygier (1993). An example of a content or predictive conditional is (39), below. (39) If it is sunny, she’ll go out. In the semantic structure of content or predictive conditionals, according to Sweetser (1990: 115), there exists a causal or at least an “enablement” rela- tionship, positive or negative, between protasis and apodosis. Barcelona (2006) argues that metonymy may have motivated the exten- sion directly. In order to understand this claim, we should first consider the role of epistemicity in predictive conditional constructions. In these condi- tionals, the speaker as conceptualizer is not normally in the “scope of predi- cation”, and is simply in the “ground” (Langacker 1999: 5–7, 22, 49–53). The construal of the situation is thus objective rather than subjective (Lan- gacker 1999: 6, 1987, 1991: Ch. 12). In other words, the situation is not pri- marily presented as conceptualized by the speaker. However, the speak- er/conceptualizer, as part of the ground, is always present, and he/she neces- sarily takes a stance towards the validity of the causal or enabling connection between protasis and apodosis. Such a stance is, in Fillmore’s (1990a, 1990b) terms, an “epistemic stance”, and it simply remains implicit in the normal understanding of predictive conditionals. In epistemic conditionals, however, it is presented as being in the conceptual foreground. The highlighting and consequent prominence of the speaker’s epistemic stance can be claimed to be due to metonymy. The CAUSAL CONNECTION BETWEEN A HYPOTHETICALLY SATISFIED CONDITION AND A RESULT acti- vates its SALIENT CONCOMITANT SUB-RELATION, namely the BUILT-IN EPIS- TEMIC CONDITIONAL CONNECTION BETWEEN SATISFIED CONDITION AND RESULT. To put it in simple terms: The causal connection between condition and result metonymically activates their built-in epistemic connection. This would then be a WHOLE FOR PART metonymy. The assertion of the causal link between the hypothetical satisfaction of a condition and the production 36 of the result includes an implicit premise-conclusion connection, which can in principle be read into any predictive conditional. An approximate para- phrase of (39) is “If (it is a true fact that) it is sunny (later), then (I can con- clude that) this fact will cause or enable her to decide to go out”. The words highlighting the implicit premise-conclusion connection are in brackets. Once the epistemic domain becomes prominent in the meaning of a con- ditional sentence, the sequence of verb forms is free from the constraints required by a predictive meaning (Sweetser 1996, and Fillmore 1990a, b). This conventional epistemic conditional meaning is paired to a set of con- ventional forms (present-present, perfect-perfect, past-past, etc.), resulting in a number of conventional epistemic conditional constructions. Sentence (38) exhibits one of these forms (perfect-perfect). 7 Metonymy in discourse 7.1 Survey The role of metonymy in pragmatic inferencing and in discourse modes is generally recognized in cognitive linguistics today. One of the first scholars to recognize the role of metonymy in inferencing was Lakoff (1987: 78–79), in his brief comments on the role of metonymy in the conversational conven- tions of Ojibwa and of English. Since then research on metonymy and prag- matic inferencing has grown remarkably. These are some of its main areas:  The metonymic basis of implicatures motivating metaphorical lexi- cal extension in grammaticalization, a famous example being the development of the future meaning of be going to (Hopper & Traugott 1993, Heine, Claudi & Hünnemeyer 1991, Traugott &  Dasher 2002). The metonymic basis of indirect speech acts (Brdar-Szabó 2009, Panther & Thornburg 1998, Thornburg & Panther 1997, Panther &  Thornburg 2003a, Pérez Hernández 2005). The metonymic basis of other types of pragmatic inferencing, in- cluding implicatures and explicatures (Barcelona 2002b, 2003b, 2005a, 2005b, 2007, 2009a, in preparation, Panther & Thornburg 2003a). Barcelona (2005a, 2005b, 2007 and in preparation) has stud-  ied metonymic chaining in implicature derivation. The role of metonymy in other aspects of discourse understanding (Gibbs 1994: Chapter 7), such as conceptual tautologies, mental re- construction of texts, euphemisms, and film, drama and artistic con- ventions and techniques, including literary styles. The use of meton- ymy in literary discourse has been the object of studies by Kuzniak 37 (2005), Dzeren-Glówacka (2005), Pluciennik (2005), and Pankhurst  (1999), among others. The function of metonymy in aphasia, iconic gestures and in sign language discourse has been discussed by Ciepiela (2005), Kwiat- kowska (2005) and Wilcox (2004), among others. Two simple examples from two of these areas are offered in this sub-section. A more complex example is offered in section 7.2. A. American Sign Language (ASL) Wilcox (2004: 203) discusses a creative ASL sign (TO-OPEN COMPUTER) occurring in the speech of a deaf ASL user. The ASL user evokes (metonymically) with his hands at his forehead the lifting up of the cover of a computer; this constitutes the metonymic sign TO-OPEN (computer cover); a laptop is a container. This sign in turn inter- acts with the conceptual metaphor THE MIND IS A CONTAINER, so that the ASL user manages to activate the notion ALLOWING ACCESS TO ONE’S PRI- VATE THOUGHTS. With that metonymic sign he communicates his willing- ness to permit public access to his private thoughts. B. Indirect speech acts Thornburg & Panther (1997) have insightfully investigated the metonymic basis of indirect speech acts. They assume the existence of “Speech Act Sce- narios”, derived from the Action Scenario, which has these parts: The BE- FORE (preconditions); the CORE (properties defining the action); its RE- SULT (the immediate outcome of a successful performance of the action); and the AFTER (intended or unintended consequences of the action which are not its immediate result). An example of a speech act scenario is the Scenario for Directive Speech Acts (S= Speaker, H = Hearer, A = action requested): (i) the BEFORE: H can do A S wants H to do A (ii) the CORE: S puts H under a (more or less strong) obligation to do A (iii) the RESULT: H is under an obligation to do A (H must / should / ought to do A) (iv) the AFTER: H will do A Metonymy is claimed to operate on these scenarios to yield indirect speech acts. These are just a few authentic examples from Thornburg and Panther’s corpus: 38 (a) A BEFORE component for the whole scenario. In (40), the meton- ymy ABILITY TO PERFORM AN ACTION FOR A LINGUISTIC ACTION (a manifestation of the metonymy POTENTIALITY FOR ACTUALITY) mo- tivates an indirect offer: (40) ‘Don’t be absurd, darling, I can advance you any amount you ask for,’ said Caroline. ‘Don’t you know I am a very wealthy woman?’ (b) A CORE/RESULT component for the whole scenario. The meton- ymy OBLIGATION TO PERFORM AN ACTION FOR A LINGUISTIC AC- TION (a manifestation of EFFECT FOR CAUSE) motivates an indirect suggestion in (41) and an indirect request in (42): (41) Julie, you’re wet. You must change. (42) Do sit down, both of you. You must certainly play for us, if you will, Mr. er.. (c) An AFTER component for the whole scenario. The metonymy A FUTURE ACTION FOR A LINGUISTIC ACTION motivates the indirect request in (43): (43) ‘Oh, Rachel, don’t you see, I can ask you to marry me now,’ he said huskily. ‘You do care for me, dearest? You will say “yes”?’ 7.2 A more complex example: Metonymy and implicature in a parlia- mentary repartee This conversation is reported to have taken place in the Spanish Parliament in the 1930’s: (44) Opposition M.P. (referring to the Prime Minister): ‘But what can we expect, after all, of a man who wears silk underpants?’ Prime Minister (rising calmly): ‘Oh, I would have never thought the Right Honorable’s wife could be so indiscreet!’ In Barcelona (2003b), I carried out a detailed study of the metonymic basis of the complex pattern of pragmatic inferences invited by this exchange. I found all of them to be guided by metonymy. These are all the inferences that were analyzed: a) Meant and conveyed by the opposition M.P. 1. The Prime Minister is a homosexual. 2. The Prime Minister is unfit for office. 39 b) Meant and conveyed by the Prime Minister. 3. The M.P.’s wife shares a secret with the Prime Minister. 4. She has told the M.P. the secret. 5. She knows that the secret consists in the fact that the Prime Minister always wears silk underpants. 6. She has seen the Prime Minister undress. 7. She has had a sexual affair with the Prime Minister, and is, thus, an adulteress (main inference). c) Perhaps meant and perhaps conveyed by the Prime Minister. 8. The Prime Minister is, after all, despite his supposed dressing habits, not a homosexual. 9. The M.P. is a cuckolded husband. 10. The M.P. knew that he was a cuckolded husband before uttering his words. 11. Through his words, the M.P. has publicly admitted that the Prime Minister is not a homosexual. 12. Through his words, the M.P. has publicly admitted that he is a cuckolded husband. 13. Through his words, the M.P. has publicly shown himself to be an utter fool. Attardo (1990) claims that some type of “frame adjustment” takes place in the comprehension of jokes and funny anecdotes. I found that the two types of frame adjustment operating in this case are “frame overlap” and “frame shift”: the UNDERWEAR frame overlaps with the HOMOSEXUALITY frame; the latter overlaps with the DISCRETION frame, which in turn overlaps and finally shifts to the HETEROSEXUALITY and ADULTERY frames. These adjustments are motivated by the above inferences, which are in turn guided by metonymy. For the sake of brevity, only the metonymic basis of inferences 3, 5, 6 and 7 is presented below:  Inference 3 is guided by RESULT (being discreet/indiscreet) FOR  CONDITION (knowing a secret). Inference 5 is guided by ENTITY (a propositional entity, namely the fact that Prime Minister uses silk underpants) FOR ONE OF ITS CON-  VENTIONAL PROPERTIES (being secret). Inference 6 is guided by FACT (knowing the underwear used by someone) FOR ONE OF ITS CONVENTIONAL EXPLANATIONS (having  seeing that person undress). The first part of inference 7 is also guided by FACT (having seeing someone undress) FOR ONE OF ITS CONVENTIONAL EXPLANATIONS (having had a sexual encounter with that person). 40  The second part of inference 7 is guided by DEFINITION (a married woman having a sexual encounter with a man not her husband) FOR DEFINED (adultery). 8 Conclusions A rich amount of evidence has been provided to: (a) illustrate the important role of metonymy in grammar and discourse. (b) emphasize the often overlooked fact that metonymy also regularly operates under (phonology, morphemics), above (phrases, clauses, sentences, utterance and discourse) the lexicon and outside oral lan- guage (i.e. ASL, gestures, etc.), and that it is primarily a conceptual mechanism. This conceptual nature explains why we often find the same general types of conceptual metonymies (EFFECT FOR CAUSE, PART FOR WHOLE, RESULT FOR CONDITION, ENTITY FOR SALIENT PROPERTY, etc.) operating at very different linguistic levels and in very different expressive and communicative modes. Some evidence has also been offered for the claim that metonymies, on the other hand, often chain to each other and to other conceptual factors, such as metaphors, or to other types of contextual information, to guide the processing of discourse (Barcelona 2005a, 2005b, 2007, 2009a, in prepara- tion). References Alm-Arvius, Christina. 2003. Figures of Speech. Lund: Studentlitteratur. Attardo, Salvatore. 1990. ‘The Violation of Grice’s Maxims in Jokes’, in Hall, Kira, Jean Pierre Koenig, Michael Meacham, Sondra Reinman & Laurel A. Sutton (eds), Proceedings of the 16th Annual Meeting of the Berkeley Linguistics So- ciety. Berkeley, CA: Berkeley Linguistics Society, 355–362. Barcelona, Antonio. In preparation. On the Pervasive Role of Metonymy in Con- structional Meaning and Structure in Discourse Comprehension: An empirical study from a cognitive-linguistic perspective. (Provisional title.) Cognitive Lin- guistics Research. Berlin: Mouton de Gruyter. —. 2011. ‘Reviewing the Properties and Prototype Structure of Metonymy’, in Benczes, Barcelona & Ruiz de Mendoza (2011): 7–57. —. 2009a. ‘Motivation of Construction Meaning and Form: The roles of metonymy and inference’, in Panther, Thornburg & Barcelona (2009): 363–401. —. 2009b. ‘Partitive Restrictive Modification of Names in English: Arguments for the metonymic motivation’, in Quaderns de Filologia de la Universitat de Va- lencia. Estudis Lingüístics 14: 33–56. 41 —. 2008. ‘The Interaction of Metonymy and Metaphor in the Meaning and Form of “Bahuvrihi” Compounds’, in Annual Review of Cognitive Linguistics 6: 208–281. —. 2007. ‘The Role of Metonymy in Meaning at Discourse Level: A case study’, in Radden et al. (2007): 51–75. —. 2006. ‘On the Conceptual Motivation of Epistemic Conditionals: Metonymy or metaphor?’, in Benczes, Reka & Szilvia Csabi (eds), The Metaphors of Sixty. Papers Presented on the Occasion of the 60th Birthday of Zoltán Kövecses. Bu- dapest: Eötvos Loránd University, 39–47. —. 2005a. ‘The Multilevel Operation of Metonymy in Grammar and Discourse with Particular Attention to Metonymic Chains’, in Ruiz de Mendoza Ibáñez, Fran- cisco José & Sandra Peña Cervel (eds), Cognitive Linguistics: Internal dynamics and interdisciplinary interaction. Cognitive Linguistics Research. Berlin: Mou- ton de Gruyter, 313–352. —. 2005b. ‘The Multilevel Role of Metonymy in Grammar and Discourse: A case study’, in Kosecki (2005b): 103–131. —. 2004. ‘Metonymy behind Grammar: The motivation of the seemingly “irregular” grammatical behavior of English paragon names’, in Radden, Günter & Klaus- Uwe Panther (eds), Motivation in Grammar. Amsterdam: John Benjamins, 357– 374. —. 2003a. ‘Metonymy in Cognitive Linguistics. An Analysis and a few Modest Proposals’, in Cuyckens, Hubert, Klaus-Uwe Panther & Thomas Berg (eds), Motivation in Language: Studies in honor of Günter Radden. Amsterdam: John Benjamins, 223–255. —. 2003b. ‘The Case for a Metonymic Basis of Pragmatic Inferencing: Evidence from jokes and funny anecdotes’, in Panther & Thornburg (2003b): 81–102. —. 2003c. ‘La Gramática de los Nombres Propios: Un viaje con retorno (por buenos motivos)’, in White, Michael, Honesto Herrera & Cristina Alonso (eds), Cogni- tive Linguistics in Spain at the Turn of the Century (II). (Metaphor and Meton- ymy) / La Lingüística Cognitiva en España en el Cambio de Siglo (II). (Metáfo- ra y Metonimia). Madrid: Universidad Autónoma de Madrid, 7–30. —. 2003d. ‘Names: A metonymic return ticket’, in Jezikoslovlje 4(1): 11–41. —. 2002a. ‘Clarifying and Applying the Notions of Metaphor and Metonymy within Cognitive Linguistics: An update’, in Dirven & Pörings (2002): 207–277. —. 2002b. ‘On the Ubiquity and Multiple-level Operation of Metonymy’, in Lewan- dowska-Tomaszczyk, Barbara & Kamila Turewicz (eds), Cognitive Linguistics Today. (Łódź Studies in Language). Frankfurt am Main: Peter Lang, 207–224. —. 2000a. ‘On the Plausibility of Claiming a Metonymic Motivation for Conceptual Metaphor’, in Barcelona (2000b): 31–58. Barcelona, Antonio (ed.). 2000b. Metaphor and Metonymy at the Crossroads. Cog- nitive Approaches. Berlin: Mouton de Gruyter. Benczes, Réka. 2006. Creative Compounding in English. The Semantics of Meta- phorical and Metonymical Noun-Noun Combinations. (Human Cognitive Pro- cessing 19). Amsterdam: John Benjamins. Benczes, Réka, Antonio Barcelona & Francisco-José Ruiz de Mendoza Ibáñez (eds). 2011. Defining Metonymy in Cognitive Linguistics. Towards a consensus view. (Human Cognitive Processing 28). Amsterdam: John Benjamins. Bierwiaczwonek, Boguslaw. 2005. ‘On Formal Metonymy’, in Kosecki (2005b): 43–67. Brdar, Mario. 2007a. ‘When Zidane is not simply Zidane, and Bill Gates is not just Bill Gates: Some thoughts on the construction of metaphtonymic meanings’, in Radden et al. (2007): 125–142. 42 —. 2007b. Metonymy in Grammar. Towards Motivating Extensions of Grammatical Categories and Constructions. Osijek (Croatia): Josip Juraj Strossmayer Uni- versity. Brdar-Szabó, Rita. 2009. ‘Metonymy in Indirect Directives: Stand-alone condition- als in English, German, Hungarian, and Croatian’, in Panther, Thornburg & Barcelona (2009): 323–336. Ciepiela, Kamila. 2005. ‘Metonymy in Aphasia’, in Kosecki (2005b): 199–208. Clark, Herbert H. 1977. ‘Bridging’, in Johnson-Laird, Philip Nicholas & Peter Cathcart Wason (eds), Thinking: Readings in cognitive science. Cambridge: Cambridge University Press, 411–420. Dancygier, Barbara. 1993. ‘Interpreting Conditionals: Time, knowledge and causa- tion’, in Journal of Pragmatics 19: 403–404. Dirven, René. 1999. ‘Conversion as a Conceptual Metonymy of Event Schemata’, in Panther & Radden (1999): 275–287. Dirven, René & Ralf Pörings (eds). 2002. Metaphor and Metonymy in Comparison and Contrast. Berlin: Mouton de Gruyter. Dirven, René & Marjolijn Verspoor. 1998. Cognitive Exploration of Language and Linguistics. (Cognitive Linguistics in Practice). Amsterdam: John Benjamins. Dzeren-Glówacka, Silwia. 2005. ‘Beating up Intelligence: Metonymy in Terry Pratchet’s novels’, in Kosecki (2005b): 335–348. Emmott, Catherine. 1998. ‘Embodied in a Constructed World: Narrative processing, knowledge representation, and indirect anaphora’, in van Hoek, Karen, Andrej A. Kibrik, & Leo Noordman (eds), Discourse Studies in Cognitive Linguistics. Amsterdam: John Benjamins, 5–29. Fauconnier, Gilles. 1997. Mappings in Thought and Language. Cambridge: Cam- bridge University Press. Fauconnier, Gilles & Mark Turner. 2002. The Way We Think. Conceptual Blending and the Mind’s Hidden Compleбities. New York: Basic Books. Feyaerts, Kurt. 1999. ‘Metonymic Hierarchies. The Conceptualization of Stupidity in German Idiomatic Expressions’, in Panther & Radden (1999): 309–334. Fillmore, Charles. 1990a. ‘Epistemic Stance and Grammatical Form in English Conditional Sentences’, in Ziolkowski, Michael (ed.), Papers from the Twenty- Sixth Regional Meeting of the Chicago Linguistics Society (I). Chicago: Univer- sity of Chicago Press, 137–162. —. 1990b. ‘The Contribution of Linguistics to Language Understanding’, in Bocaz, Aura (ed.), Proceedings of the First Symposium on Cognition, Language and Culture. Santiago: Universidad de Chile, 109–128. —. 1985. ‘Frames and the Semantics of Understanding’, in Quaderni di Semantica 6(2): 222–254. Geeraerts, Dirk & Yves Peirsman. 2011. ‘Zones, Facets and Prototype-based Me- tonymy’, in Benczes, Barcelona & Ruiz de Mendoza (2011): 88–102. Gibbs, Raymond W. Jr. 2007. ‘Experiential Tests of Figurative Meaning Construc- tion’, in Radden et al. (2007): 19–32. —. 1999. ‘Speaking and Thinking with Metonymy’, in Panther & Radden (1999): 61–76. —. 1994. The Poetics of Mind. Figurative Thought, Language, and Understanding. Cambridge: Cambridge University Press. Goldberg, Adele. 2006. Constructions at Work: The nature of generalization in language. Oxford: Oxford University Press. Greenbaum, Sydney & Randolph Quirk. 1990. The Student’s Grammar of the English Language. London: Longman. 43 Heine, Bernd, Ulrike Claudi & Friederike Hünnemeyer. 1991. Grammaticalization. A Conceptual Framework. Chicago: University of Chicago Press. Hilpert, Martin. 2007. ‘Chained Metonymies in Lexicon and Grammar: A cross- linguistic perspective on body terms’, in Radden et al. (2007): 77–98. Hopper, Paul & Elizabeth Closs Traugott. 1993. Grammaticalization. Cambridge: Cambridge University Press. Jones, Daniel. 1969 [1918]. An Outline of English Phonetics. 9th ed. Cambridge: Heffer. Kosecki, Krzyscztof. 2005a. ‘Some Remarks on Metonymy in Compounding’, in Kosecki (2005b): 241–251. Kosecki, Krzyscztof (ed.). 2005b. Perspectives on Metonymy. Proceedings of the International Conference ‘Perspectives on Metonвmв’, Held in Lяdг, Poland, May 6–7, 2005. Frankfurt am Main: Peter Lang. Kövecses, Zoltán & Günter Radden. 1998. ‘Metonymy: Developing a cognitive linguistic view’. Cognitive Linguistics 9(1): 37–77. Kuzniak, Marek. 2005. ‘Part-whole Relations in the Selected Epigrams by J. Staudynger’, in Kosecki (2005b): 323–333. Kwiatkowska, Alina. 2005. ‘Pre-linguistic and Non-linguistic Metonymy’, in Kosecki (2005b): 297–307. Lakoff, George. 1987. Women, Fire and Dangerous Things. What Categories Re- veal About the Mind. Chicago: University of Chicago Press. Lakoff, George & Mark Johnson. 1980. Metaphors We Live By. Chicago: University of Chicago Press. Langacker, Ronald W. 2009. ‘Metonymic Grammar’, in Panther, Thornburg & Bar- celona (2009): 45–71. —. 1999. Grammar and Conceptualization. Berlin: Mouton de Gruyter. —. 1993. ‘Reference-point Constructions’, in Cognitive Linguistics 4: 1–38. —. 1991. Concept, Image and Symbol. The Cognitive Basis of Grammar. Berlin: Mouton de Gruyter. —. 1987. Foundations of Cognitive Grammar. Vol. 1: Theoretical Prerequisites. Stanford: Stanford University Press. Michaelis, Laura. 2004. ‘Type Shifting in Construction Grammar: An integrated approach to aspectual coercion’, in Cognitive Linguistics 15: 1–67. ‘Mount Bethel’. 1996. On-line at: http://hikingincolorado.org/beth.html (retrieved 15 August 2008). Nathan, Geoffrey. 1996. ‘Steps Towards a Cognitive Phonology’, in Hurch, Bern- hard & Richard Rhodes (eds), Natural Phonology: The state of the art. Berlin: Mouton de Gruyter, 107–120. —. 1986. ‘Phonemes as Mental Categories’, in Nikiforidou, Vassiliki, Mary Van Clay, Mary Niepokuj & Deborah Feder (eds), Proceedings of the Annual Meeting of the Berkeley Linguistics Society 12. Berkeley: Berkeley Linguistics Society, 212–223. Palmer, Gary, Russell S. Rader & Art D. Clarito. 2009. ‘The Metonymic Basis of a “Semantic Partial”: Tagalog lexical constructions with ka-’, in Panther, Thorn- burg & Barcelona (2009): 111–144. Pang, Kam-Yiu. 2006. A Partitioned-Narrative Model of the Self: Its linguistic man- ifestations, entailments, and ramifications. Ph. D. dissertation. Dunedin: Uni- versity of Otago, New Zealand. Pankhurst, Anne. 1999. ‘Recontextualization of Metonymy in Narrative and the Case of Morrison’s Song of Solomon’, in Panther & Radden (1999): 385–399. Panther, Klaus-Uwe & Günter Radden (eds). 1999. Metonymy in Language and Thought. Amsterdam: John Benjamins. 44 Panther, Klaus-Uwe & Linda Thornburg. 2003a. ‘Introduction: On the nature of con- ceptual metonymy’, in Panther & Thornburg (2003b): 1–20. —. 2002. ‘The Roles of Metaphor and Metonymy in English -er Nominals’, in Dirven & Pörings (2002): 279–319. —. 2000. “The EFFECT FOR CAUSE Metonymy in English Grammar”, in Barcelona (2000b): 215–231. —. 1999. ‘The Potentiality for Actuality Metonymy in English and Hungarian’, in Panther & Radden (1999): 333–357. —. 1998. ‘A Cognitive Approach to Inferencing in Conversation’, in Journal of Pragmatics 30: 755–769. Panther, Klaus-Uwe & Linda Thornburg (eds). 2003b. Metonymy and Pragmatic Infer- encing. (Pragmatics and Beyond New Series). Amsterdam: John Benjamins. Panther, Klaus-Uwe, Linda Thornburg & Antonio Barcelona (eds). 2009. Metonymy and Metaphor in Grammar. (Human Cognitive Processing 25). Amsterdam: John Benjamins. Papafragou, Anna. 1996. ‘On Metonymy’, in Lingua 99: 169–195. Paradis, Carita. 2004. ‘Where does Metonymy Stop? Senses, Facets and Active Zones’, in Metaphor and Symbol 19(4): 245–264. Pérez Hernández, Lorena. 2005. ‘High-level Metonymies in the Understanding of Modality’, in Kosecki (2005b): 133–146. Pluciennik, Jaroslaw. 2005. ‘Princess Antonomasia, Individualism, and the Quixo- tism of Culture: A case of “Tristram Shandy” by Laurence Sterne’, in Kosecki (2005b): 349–366. Pustejovsky, James. 1995. The Generative Lexicon. Cambridge, Mass: MIT Press. Radden, Günter. 2009. ‘Generic Reference in English: A metonymic and conceptual blending analysis’, in Panther, Thornburg & Barcelona (2009): 199–228. —. 2005. ‘The Ubiquity of Metonymy’, in Otal Campo, José Luis, Ignasi Navarro i Ferrando & Begoña Bellés Fortuño (eds), Cognitive and Discourse Approaches to Metaphor and Metonymy. Castelló (Spain): Universitat Jaume I, 11–28. —. 2000. ‘How Metonymic are Metaphors?’, in Barcelona (2000b): 93–108. Radden, Günter & René Dirven. 2007. Cognitive English Grammar. Amsterdam: John Benjamins. Radden, Günter, Klaus-Michael Köpcke, Thomas Berg & Peter Siemund (eds). 2007. Aspects of Meaning Construction. Amsterdam: John Benjamins. Radden, Günter & Zoltán Kövecses. 1999. ‘Towards a Theory of Metonymy’, in Panther & Radden (2000): 17–59. Ruiz de Mendoza Ibáñez, Francisco José & José Luis Otal Campo. 2002. Metonymy, Grammar and Communication. (Colección Estudios de Lengua Inglesa 7). Al- bolote (Granada): Comares. Ruiz de Mendoza Ibáñez, Francisco José & Lorena Pérez Hernández. 2001. ‘Metonymy and the Grammar: Motivation, constraints and interaction’, in Language and Communication 21(4): 321–357. Ruiz de Mendoza Ibáñez, Francisco José & Ricardo Mairal Usón. 2007. ‘High-level Metaphor and Metonymy in Meaning Construction’, in Radden et al. (2007): 33–49. Soloshenko, Alexander. 2005. ‘Metaphor and Metonymy in Biblical Proper Names’, in Kosecki (2005b): 309–321. Sweetser, Eve. 1996. ‘Mental Spaces and the Grammar of Conditional Construc- tions’, in Fauconnier, Gilles & Eve Sweetser (eds), Spaces, Worlds and Gram- mars. Chicago: The University of Chicago Press, 318–333. —. 1990. From Etymology to Pragmatics: Metaphorical and cultural aspects of semantic structure. Cambridge: Cambridge University Press. 45 Szawerna, Michal. 2005. ‘Deverbal Nominalization as a Type of Metonymic Exten- sion from Processes to Things’, in Kosecki (2005b): 147–155. Taylor, John. 1995 [1989]. Linguistic Categorization. Prototypes in Linguistic Theo- ry. Oxford: Oxford University Press. Thornburg, Linda & Klaus-Uwe Panther. 1997. ‘Speech Act Metonymies’, in Liebert, Wolf-Andreas, Gisela Redeker & Linda Waugh (eds), Discourse and Perspec- tive in Cognitive Linguistics. (Current Issues in Linguistic Theory 151). Am- sterdam: John Benjamins, 205–219. Traugott, Elizabeth Closs & Richard B. Dasher. 2002. Regularity in Semantic Change. (Cambridge Studies in Linguistics 96). Cambridge: Cambridge Univer- sity Press. Turner, Mark & Gilles Fauconnier. 2000. ‘Metaphor, Metonymy and Binding’, in Barcelona (2000b): 133–145. —. 1995. ‘Conceptual Integration and Formal Expression’, in Metaphor and Sym- bolic Activity 10: 183–204. Warren, Beatrice. 2002. ‘An Alternative Account of the Interpretation of Referential Metonymy and Metaphor’, in Dirven & Pörings (2002): 113–130. —. 1999. ‘Aspects of Referential Metonymy’, in Panther & Radden (1999): 121– 135. Wilcox, Phyllis P. 2004. ‘A Cognitive Key: Metonymic and metaphorical mappings in ASL’, in Cognitive Linguistics 15(2): 197–222. Ziegeler, Debra. 2007. ‘Arguing the Case against Coercion’, in Radden et al. (2007): 99–123. 46 Gerard Steen When is Metaphor Deliberate? Abstract. Contemporary metaphor research sees metaphor as conventional, auto- matic and unconscious. This raises the question of when metaphors are used deliber- ately. It appears that all novel metaphors and some conventional metaphors are used deliberately, and that this happens when they are intended to produce a change in the perspective of the addressee on the topic of a stretch of discourse. This typically happens for all metaphors that are expressed directly, such as extended non-literal comparisons and similes. Indirect metaphors can also be used deliberately, but this does not occur very frequently. The paper explains how these factors of metaphor interact with each other to produce deliberate or non-deliberate metaphor. Keywords: metaphor deliberateness, linguistic form, conceptual structure, commu- nicative function 1 Introduction When Barack Obama addressed the nation as the next president of the Unit- ed States of America in November 2008, he gave a heavily rhetorical speech which included a good deal of metaphor. Some of those metaphors were clearly deliberate but others belong to the common register of political speeches or even the contemporary variety of American English; the latter may have simply been used without any awareness of their metaphorical nature on the part of the speaker (or his speech writers). In this essay I will use some of Obama’s metaphors to reflect on the nature of the difference between deliberate and non-deliberate metaphor, a distinction which has been ignored in metaphor research for far too long. For instance, the recent Cambridge Handbook of Metaphor and Thought does not have ‘deliberate metaphor’ in its index (Gibbs 2008). I will therefore also relate the distinc- tion between deliberate and non-deliberate metaphor to the current views of metaphor as a regular ingredient of language and thought (cf. Steen 2008). Obama began his speech as follows: If there is anyone out there who still doubts that America is a place where all things are possible; who still wonders if the dream of our founders is alive in our time; who still questions the power of our democracy, tonight is your an- swer. 47 This first utterance contains a number of metaphorically used words, of which dream is probably the most conspicuous. It harks back to Dr Martin Luther King’s famous “I have a dream” from 1963. I have no doubt that the allusion is deliberate, but I am not sure whether it follows that its metaphori- cal nature is also deliberately invoked. That is an issue to which I will return. The word dream is metaphorically used because its meaning in this par- ticular utterance (its ‘contextual meaning’) is ‘something good that you hope you will have or achieve in the future’ (sense description 2 in the Macmillan English Dictionary for Advanced Learners, Rundell 2002). This contextual meaning is distinct from the more basic meaning of the word dream, which can be defined as ‘something that you experience in your mind while you are sleeping’ (Macmillan, sense 1): something that happens during your sleep is decidedly distinct from something you hope for. If the contextual meaning of a word is distinct from its basic meaning, but can be understood by compari- son with it, the word is metaphorically used (Pragglejaz Group 2007). Since one function of dreams is to show you situations that you might hope for, the ‘hope’ sense can be argued to be distinct from but comparable to the ‘sleep’ sense of dream. Another potentially metaphorically used word is alive. The Longman Dic- tionary of Contemporary English shows that its contextual sense is number 2, ‘still existing, continuing to exist’. This is distinct from sense 1, ‘not dead, still living and not dead’, which may be regarded as the word’s basic sense. The example sentences for the contextual sense show that it typically applies to abstract entities, including ‘ancient traditions’, ‘Christianity’, and ‘the sport’. That is also how it is used here, because it is applied to the abstract entity ‘dream’ in the sense of ‘hope’, not the concrete embodied experience ‘dream’. By contrast, all illustrations of the basic sense of alive are con- cerned with people. As a result, it may be claimed that the contextual sense of alive (‘still existing’) can be understood as distinct from but in compari- son with the basic sense (‘not dead’): alive is hence also used metaphorical- ly. Again, whether it is used deliberately as a metaphorical expression is another matter, to which I will return. I have been concerned with the identification of metaphor in grammar and usage for more than ten years (Steen 2007). My quest started at a conference of the Poetics And Linguistics Association (PALA) in 1998, where I pre- sented a paper on listeners’ responses to some 40 metaphorically used words in a song by Bob Dylan, ‘Hurricane’ (Steen 2002a, 2004). I employed a range of metaphorically used words as my stimuli to examine how often the participants in the study recognized them as metaphorical, and received criti- cal questions and comments from the audience regarding my grounds for analytically identifying some of them as metaphorical. The discussion with the audience revealed that, in spite of the abundant cognitive-linguistic re- search on the ubiquity of metaphor in language (cf. Kövecses 2002 for an 48 introduction), there were serious problems with the reliable identification of metaphor in any stretch of natural discourse. Since this difficulty jeopardizes the validity of any empirical study of metaphor in language, a small group of conference attendees decided to ex- plore the possibility of developing a procedure for metaphor identification in literary and nonliterary language use. This led to the foundation of a PALA Special Interest Group for metaphor, which published its findings in a spe- cial issue of the society’s journal, Language and Literature (Crisp 2002; Crisp, Heywood, & Steen 2002; Heywood, Semino, & Short 2002; Steen 2002b). The publications of this group amounted to a conceptual framework within which methodological research on the practical identification of met- aphor in discourse could be carried out. This led to the formation of another group of metaphor researchers, the Pragglejaz Group, which included three members of the original PALA group (Peter Crisp, Elena Semino and myself). The Pragglejaz Group was much broader, however, and also included cognitive linguists (Alan Cienki, Joe Grady, Zoltán Kövecses), applied linguists (Lynne Cameron, Alice Deignan, Graham Low), and a psycholinguist (Ray Gibbs). A series of meet- ings were held between 2000 and 2006 in the form of annual three-day workshops where we collaborated to develop an operational definition and procedure that would be acceptable across the board in the various disci- plines represented in the group. The results of this work were eventually published in the form of a metaphor identification procedure, or MIP (Prag- glejaz Group 2007). The procedure was shown to be reliable by statistical testing, and has since been adopted by many metaphor scholars in their prac- tical work. While the Pragglejaz Group were preparing their publication, I started a third research group at VU University Amsterdam, a group which aimed to apply MIP to a substantial amount of discourse data. Since September 2005, two research projects, involving a total of seven PhD students, have been carried out, annotating a number of modest corpora in English and Dutch for their use of metaphor. We examined 50,000-word samples of fiction, sci- ence, news, and conversation from the BNC-Baby, and samples of the same size from Dutch news texts and conversations. Our application of the Prag- glejaz Group’s MIP procedure led to the identification of a number of new difficulties (Steen, Biernacka, et al. 2010). We therefore developed our own refinement and extension, labeled MIPVU, in which we have proposed our own solutions (Steen, Dorst, et al. 2010). Our own procedure has greater statistical reliability than MIP, as well as broader coverage of metaphor in discourse. Details of its structure and scope have been motivated by an ex- tensive methodological analysis of theory and research on metaphor across linguistics (Steen 2007). Our development and application of MIPVU to our large samples of data has recently led to a new focus in my research, the deliberate use of meta- 49 phor, which is the topic of the present paper (Steen 2008). In what follows I will sketch out how our work on metaphor identification has led to a distinc- tion between direct and indirect metaphor, which in turn has drawn my atten- tion to the deliberate versus non-deliberate use of any form of metaphor. The Pragglejaz Group’s MIP procedure is focused on the identification of indi- rect metaphor, and I will briefly explain some of its main properties in the next section. Indirect metaphor will be distinguished from direct metaphor in section 3. The analysis of our various corpora by MIPVU has subsequently led to new insights into the varying communicative status of diverse forms of metaphor, with a special role for deliberate metaphor; this will be dis- cussed in section 4. Some implications of this analysis will be presented in the final section. 2 Metaphor identification The basis of the identification of metaphor in any spoken or written message is the assumption that it can be distinguished from words that are used di- rectly. Thus, when Obama starts out on his speech, most of his words are used directly, in some specific sense of the term ‘direct’, to be explained here: If there is anyone out there who still doubts that America is a place where all things are possible; who still wonders if the dream of our founders is alive in our time; who still questions the power of our democracy, tonight is your an- swer. Most words in this utterance can be assumed to evoke a corresponding con- cept in the mental encyclopedia of the typical language user that in turn can be directly connected to the referent that the language user is supposed to set up in their mental representation of the utterance. Obama directly posits a person out there who performs the action of having a doubt about something, and that something is concerned with America as being a place where all things are possible. There is not one word or one concept in this stretch of discourse that requires any special action when it comes to setting up their intended referents: these are all directly related to the corresponding con- cepts and words of the utterance. When we turn to the next part of the utterance, however, the picture changes. For it can be argued that both dream and alive are not used directly. They have a basic sense which has to do with an embodied experience (a real dream) and life in some human or biological sense (alive) which does not apply in this context. If we start out from the basic sense of these words, or so the cognitive-linguistic argument goes, they activate a concept which is not part of the situation that is intended by the speaker, for he is not speaking 50 about sleep or biology. As a result, it is assumed that the concepts triggered by the basic sense of the words should be used to project another referent, which is derived by a metaphorical mapping. This is an illustration of the cognitive-linguistic argument in current metaphor research that the meaning of metaphor in discourse is based on cross-domain mapping. Since meta- phorical mappings set up correspondences across conceptual domains, for instance between dreams and hopes, they are based on some form of similar- ity, which in fact may take all sorts of shapes (Steen 2007). This argument takes seriously the claim in cognitive linguistics that meta- phor in language is based on indirect as opposed to direct meaning (e.g., Lakoff 1986; Gibbs 1993). When words like attack and defend are used to talk about argumentation, they are called metaphorical because there is an opposition between the presumably basic physical, concrete senses of these words on the one hand and their abstract argumentation senses on the other. This contrast is also the reason for postulating a more encompassing meta- phor in conceptual structure, or ‘thought’, which is called ARGUMENT IS WAR (Lakoff & Johnson 1980, 1999; Kövecses 2002). MIP and MIPVU capitalize on this theoretical analysis of metaphor iden- tification. They instruct discourse analysts to identify the contextual mean- ings of words and to check whether these can be contrasted with a more basic sense of that same word. If that is so, and the contrast can be resolved by comparison, the word is used metaphorically. This argument applies to dream, alive, and even to the preposition in when used by Obama in the phrase in our time: the contextual sense is abstract, temporal, while the basic sense is spatial, evoking a container. There have been numerous publications about the metaphorical conceptualization of time as space which can support the argument that this use of in belongs to this pattern. What is important about this approach to metaphor identification is that it is not specific to metaphor. The methodology is supported by theoretically and empirically supported distinctions in discourse studies. In particular, there are three different mental representations of discourse that have been generally recognized in the psychology of discourse which play a role here (e.g. Clark 1996; Kintsch 1998). We need separate descriptions of words, concepts, and referents as parts of distinct mental models of the message, since these are three distinct levels of cognitive discourse representation. Van Dijk & Kintsch (1983) have suggested that these levels can be called the surface text (involving the representation of the linguistic structure of a mes- sage, including our words), the text base (representing the conceptual struc- tures of a message in the form of a series of linearly and hierarchically relat- ed propositions, including our concepts), and the situation model (involving a more abstract representation, like a picture or a film, of the content of the message, which includes our referents). When we are examining metaphor in discourse, it is the step from the concepts in the text base to the referents in the situation model where we need additional analysis in the form of a cross- 51 domain mapping (or projection by comparison or similarity). Thus, Obama’s utterance addresses the political situation of America but then seems to de- rail because it switches to words and concepts that have to do with sleep (dream, DREAM), biology (alive, ALIVE), and space (in, IN). If these concepts were to set up their corresponding referents in the situation model, they would constitute a grave danger to its coherence. This in fact is a potent symptom of metaphor in discourse. It is the reason why some discourse ana- lysts concentrate on metaphor as incongruity (Cameron 2003; Charteris- Black 2004; Caballero 2006; Steen 2007). Metaphor involves cross-domain mapping (or ‘semantic transfer’ in Cameron’s terminology) for the same reason: the threat to coherence is resolved by making available the truly in- tended referent by cross-domain mapping, from the activated source-domain concept in the text base to a presumed target domain concept in the situation model—in our case towards hope, continuous existence, and a temporal rela- tion. What this analysis takes for granted is the fundamental status of the basic senses of words (Pragglejaz Group 2007). These are defined as related to concrete embodied human experience, on the assumption that these were the first senses of most words over the historical development of language (e.g. Sweetser 1990). There is a vast amount of metaphor research in cognitive linguistics that has adopted this assumption (cf. Gibbs 2006). This is the reason why our approach to metaphor identification has followed this route: it makes explicit what the consequences are of this cognitive-linguistic premise for the study of the use of metaphor in discourse. However, it is questionable whether such a perspective is completely ade- quate. Words may be rapidly disambiguated and activate the intended target- domain concept straightaway, in the text base, that is, so that these concepts do not require any subsequent metaphorical mapping from one concept to another (Glucksberg 2001, 2008). The metaphorical sense of many words may have become the most salient sense to the contemporary language user, so that the basic sense is not involved in the stage of concept activation (Gio- ra 2003, 2008). This has led to what I have called the paradox of metaphor: most metaphors may not be processed metaphorically, that is, by means of a cross-domain mapping from one concept to another, as is claimed by cogni- tive linguists (Steen 2008). These are problems, however, which can only be addressed when linguistic metaphor identification is connected to psycholin- guistic research on the processing of words in context, which I will leave aside for the moment. When I point out below that MIP can reliably identify metaphors in dis- course, therefore, I do not mean to suggest that all of these words are pro- cessed as metaphors, that is, by cross-domain mapping or comparison, by language users. This is a position that many cognitive linguists adopt, but my claim is different. What MIP and MIPVU do is identify words as potentially metaphorical, given a number of assumptions about contextual and basic 52 meanings. It is possible for them to be processed by comparison, but their behavior in processing is a different research area. Issues in processing do motivate our particular way of identifying metaphor in language structure but lie outside the scope of this essay (Steen 2007). Having said this, the bulk of metaphors in discourse can be identified with this simple set of assumptions and their translation into an explicit and sys- tematic procedure. To give one more sample, let us examine the following excerpt: I just received a very gracious call from Senator McCain. He fought long and hard in this campaign, and he’s fought even longer and harder for the country he loves. He has endured sacrifices for America that most of us cannot begin to imagine, and we are better off for the service rendered by this brave and selfless leader. I congratulate him and Governor Palin for all they have achieved, and I look forward to working with them to renew this nation’s promise in the months ahead. One obvious candidate for metaphor identification is the first instance of fought, where the contextual meaning of fight has to do with a political struggle, whereas the basic meaning of fight pertains to a physical struggle, and the two senses are distinct and can be compared. Another candidate would be ahead, where the spatial sense (e.g. the road ahead) can act as a basis for the temporal sense (the months ahead). These are words that with- out difficulty can be identified with the help of MIP and MIPVU as meta- phorically used. Complications arise when we turn to campaign. It is unclear whether the more general social or political sense is basic, or the military one. Both Macmillan and Longman list the first sense as the most frequent sense, and present the military sense as secondary. Historically, however, it is the mili- tary sense that comes first and is basic. This is where frequency and salience clash with historical precedence as criteria for deciding what counts as the basic sense: if the basic sense of campaign is its first sense, then it is not metaphorically used, but if it is its second sense, then it is metaphorically used. Other complications arise when we turn to long and hard. If we take these words as deriving from their corresponding adjectives, then they have basic senses which differ from their contextual senses: in that case, long has to do with space, and hard has to do with materials. This approach should classify these words as metaphorical. However, when we take these words as repre- senting a particular word class by themselves, then the adverb long only pertains to time, and the adverb hard has to do with a lot of effort. From that perspective, these words are not metaphorical. Theoretical analysis has to motivate which of the two approaches is followed, and why. The position we have adopted with MIPVU deviates from the views advocated by the Prag- glejaz Group in this respect: we argue that we are dealing with words at the 53 level of word classes that activate concepts that set up referents of a particu- lar type in a situation model (entities, processes, attributes, and so on). Therefore words should be seen in their grammatical role, to the effect that adverbs differ from adjectives, and nouns from verbs. In our analysis, long and hard would not be identified as metaphorically used. (They may have metaphorical origins in morphological conversion, but that is another matter, which does not pertain to synchronic language use.) There are a number of comparable details that analysts can have different opinions about (Steen, Dorst, et al. 2010), but I need to move on to the oppo- sition between indirect and direct metaphor. What we have seen in this sec- tion is that a reliable tool has been developed to identify the bulk of meta- phors in natural discourse. There are fundamental issues of theory and re- search involved in this work on metaphor identification, issues which may receive different answers in different research projects, but they have been clarified in a way that has stimulated further discussion and reflection. 3 Indirect versus direct metaphor One consequence of the close examination of metaphor identification is the discovery of the importance of the distinction between indirect and direct met- aphor. When we were identifying the metaphorically used words in our data, it soon became clear that there was one type of problem which was unlike all the others, of which some were briefly mentioned above. This special problem was a different class of metaphors in language, a class which could not be handled by the procedure advanced by the Pragglejaz Group. This is the phe- nomenon of direct metaphor, which we will turn to in this section. Since it does not occur in Barack Obama’s acceptance speech, I will have to resort to examples from other sources to illustrate what is going on. Consider the following excerpt from a news text that is part of BNC- Baby, the corpus from which we took our materials for analysis (Steen, Dorst, et al. 2010): IN SYSTEMS development nothing is more fundamental than assessing user requirements. (...) But many system developers are unable to assess require- ments properly. They seem to think that you can ask a businessman what his requirements are and get an answer that amounts to a draft system specifica- tion. A doctor doesn’t ask his patient аhat treatment to prescribe. The patient can explain only what the problem is. It is the doctor that provides the reme- dy. (...) A user may have a deep knowledge of business problems, but know- ing little about computers, has no idea how they should be tackled. Yet, ana- lysts are heard asking time and again, ‘Tell me what you want. (...)’ But of course the users don’t know what they want, so they end up getting another duff system. An effective analyst provides the same service to the business as the doctor provides to the patient. (A8R, my italics) 54 This text is about the problematic relation between system developers and system users, in this case, businessmen. The first three sentences in the ex- cerpt delineate this topic in language that is mostly nonmetaphorical. Then the text moves to another frame of reference. For three sentences in a row, it discusses the relationship between doctors and patients. This apparent di- gression is clearly intended to illuminate the comparable problematic rela- tionship between system developers and system users. The section on doc- tors and patients offers an alien conceptual domain from which the reader can re-view the most important aspects of the previous, dominant conceptual domain of system development, which is the true topic of the text. Since the two domains are clearly distinct but can also be understood by comparison, this is the basis of a cross-domain, i.e. metaphorical, mapping. To hammer this intention home, the last sentence in the excerpt spells out the most im- portant parts of the mapping in the text: “An effective analyst provides the same service to the business as the doctor provides to the patient.” What is important about this excerpt is that the words may be related to a metaphorical comparison, but that the language is not used indirectly. The sentences about the doctor and the patient use words, concepts and referents that are all directly related to each other. The situation model of the text momentarily needs a separate area in which the reader has to set up a new set of referents (including doctors, patients, and a number of other features) that exhibit particular relations with each other that are also explicitly named in the text. At this point, no cross-domain mapping is needed from the concept of DOCTOR or PATIENT to some other referent than ‘doctor’ or ‘patient’, for we need to understand these sentences in their own right. There is no ques- tion of a contrast between contextual and basic meanings of these words which would need to be resolved in a way comparable to what we saw when we looked at dream, alive, in, fight and ahead in the previous section. The point is that the language is used directly to set up a conceptual do- main which briefly functions as a local topic in its own right. This alien topic is then exploited to carry out a metaphorical comparison with the more en- compassing topic: system developers and users. But the metaphorical com- parison does not take place at the level of some tension between words and their meanings (intended referents) in their utterances. This is why we call this phenomenon ‘direct metaphor’. There is some conceptual domain that functions as a source in a metaphorical mapping, and it is expressed directly as such by the language. The directly expressed conceptual structure requires textual integration into the encompassing target domain, but it is an autono- mous conceptual domain which has been separately represented as such in the situation model. These are metaphor-related words, but, in contrast to dream, fight and so on, they are not metaphorically used words. To spell out the contrast in full, when Obama says that Senator McCain has fought long and hard, he does not want us to set up a situation model in which we see McCain in physical engagement with Obama himself. Instead, 55 he evokes a situation model in which McCain has spent much energy on his political activities. This situation model, however, is designated indirectly, i.e. by a word which activates a physical engagement concept (or, again, so runs the cognitive-linguistic argument). By contrast, when the systems de- velopment text mentions doctors and patients, it instead instructs us to set up a situation model in which we think about doctors and patients, at least for the duration of those three sentences. Subsequent sentences will take us back to the domain of system developers and users, and then our independent representation of doctors and patients will serve its comparative function. The identification of direct metaphor can be based on the same principles as the identification of indirect metaphor. What we need to do is to shift the criterion for metaphor identification from a contrast between contextual and basic senses of words to a contrast between textual and local functions of referents, or even topics. Thus, the local function of the doctor-patient refer- ents is to evoke the domain of doctor-patient interaction, but the textual function of these referents is to provide a source domain for a cross-domain mapping to the target domain of systems development. The technical details of this elaboration can be found in Steen, Dorst, et al. (2010). Direct metaphor does not occur very frequently. The bulk of metaphors in our corpus data, which comprises 190,000 words of English and 130,000 words of Dutch, are indirect metaphors. We estimate that direct metaphors account for at most one percent of all metaphors in discourse. It is not acci- dental that Obama’s speech does not contain any direct metaphors. Direct metaphor can be extended, as in our systems development text, or remain restricted to one sentence. Its best-known rhetorical form is the simi- le. When metaphorical mappings are expressed as similes, this often happens as an adverbial adjunct to a metaphorically used verb, as in flв like an eagle. Croft & Cruse (2004) have labelled this phenomenon metaphor-within- simile and vice versa. Nominal similes in the A is like B formula do not oc- cur much—an example could be Neil Young’s Everв junkie’s like a setting sun. This throws an interesting light on the general validity of much psycho- linguistic research, which is often concerned with the difference between simile and metaphor (e.g. Chiappe, Kennedy & Smykowski 2003; Gentner & Bowdle 2008; Glucksberg 2008). A full discussion of these other forms of metaphor is beyond the remit of this essay. Some further reflections can be found in Steen 2007, chapter 11 and in Goatly 1997. There are clear connections between direct metaphor and parable and allegory, which have been discussed by for instance Crisp (2001, 2005, 2008). For now we can conclude by saying that direct metaphor is a fundamentally different class of metaphor than most metaphor in dis- course, but one which can also be reliably identified from one and the same encompassing methodological perspective (Steen 2007; Steen, Dorst, et al. 2010). 56 4 From direct metaphor to deliberate and non-deliberate metaphor Our attention to direct metaphor, both as an empirical phenomenon in our corpus and as a methodological difficulty for our MIPVU procedure, made me realize that direct metaphor was different from indirect metaphor in yet another way. It is not just linguistically different from indirect metaphor, in that it embodies a direct expression of a conceptual domain that functions as a source in a metaphorical comparison. It is also communicatively different, because it leaves the addressee no option but to pay explicit attention to the source domain as a source domain. Indirect metaphor is deliberately meta- phorical. Deliberate metaphor is a new theme in metaphor research, which has been prepared (Goatly 1997) and announced by a number of researchers, most notably Shen & Balaban (1999), Cameron (2003; cf. Cameron 2008), Char- teris-Black & Musolff (2003), and Goddard (2004). It is to be contrasted with the major concern of mainstream metaphor research, which emphasizes that metaphor is conventional, automatic, and unconscious (cf. Lakoff & Johnson 1980, 1999; Gibbs 1994, 2008). This message has had its impact, but it appears that the time is now ripe to formulate it in a slightly more so- phisticated manner. In particular, most metaphor seems to be conventional and automatic and unconscious, but some metaphor clearly is not, so that the question arises: When is metaphor deliberate? Moreover, can we account for the reasons when metaphor is used deliberately as opposed to non- deliberately? These are the questions that focus on the communicative func- tion of metaphor, while making use of the wealth of research findings about metaphor in language and metaphor in thought (Steen 2008). One part of the answer to this question is that metaphor is deliberate when it is direct. If language users instruct their addressees to set up a separate source domain in its own right, by directly designating the referents of that source domain in their own terms, they deliberately evoke an alien concep- tual domain which functions as a basis for metaphorical comparison regard- ing the dominant conceptual domain that is the relevant topic in the dis- course. Thus, the writer of the systems development text quoted above can- not be claimed to employ metaphor non-deliberately. The text evokes a non- literal comparison between two distinct domains on purpose, and even makes that intention explicit at the end of the excerpt. The argument here is that deliberate metaphor is defined by the property that it leaves the addressee no option but to consciously set up a cross- domain mapping. This argument can be made for all direct metaphor, includ- ing simile. Since there is not a lot of direct metaphor in discourse, this would seem to suggest that there is not a lot of deliberate metaphor in discourse, either. And this in turn would accord with the emphasis on metaphor in con- 57 temporary metaphor research as typically conventional, automatic, and un- conscious. The default assumption about metaphor can hence be that it is non-deliberate, a claim that would be endorsed by most current metaphor researchers. These insights seem to work well together, and form the basis of an answer to the question as to when metaphor is deliberate. However, even though (a) metaphor is typically non-deliberate, and (b) by definition is deliberate when it is direct, it does not follow that (c) all indirect metaphor is also non-deliberate. Nor is it true that all conventional metaphor is not deliberate. People can use conventional and indirect metaphor in high- ly deliberate ways, as I shall now illustrate. This possibility clearly shows that we have three distinct dimensions of metaphor: the linguistic dimension of (in)directness, the conceptual parameter of conventionality, and the com- municative dimension of deliberateness. Instead of a two-dimensional model of metaphor in language and thought, we need a three-dimensional model of metaphor in language, thought, and communication (Steen 2008). Let us consider some examples to demonstrate how this approach may work. Here are some metaphors that are non-deliberate. They are also con- ventional and indirect—as is the typical case. To my campaign manager David Plouffe, my chief strategist David Axelrod, and the best campaign team ever assembled in the history of politics — you made this happen, and I am forever grateful for what you’ve sacrificed to get it done. If we include campaign as metaphorical because its basic sense is defined as military, then all of the words in italics are metaphorical in an indirect way, as defined by both MIP and MIPVU. None of these metaphorically used words display any signs of having been used deliberately, or consciously, as metaphors. They do not necessarily invite the addressee to set up a cross- domain mapping to understand one thing in terms of another. They all have conventionally metaphorical senses that are the intended meaning of the speaker, and if addressees were to access these meanings by immediate sense disambiguation, and then activate the corresponding target domain concepts, nothing in the utterance would have been lost in the process of communication. In the following example, the words in and this are also metaphorical, and conceptually conventional, linguistically indirect, and communicatively non- deliberate: I want to thank my partner in this journey, a man who campaigned from his heart and spoke for the men and women he grew up with on the streets of Scranton and rode with on that train home to Delaware, the Vice President- elect of the United States, Joe Biden. 58 But the situation is different for journey. Its contextual sense is also conven- tionally metaphorical and indirect. However, this metaphorical sense (‘a process of changing and developing over a period of time’) is marked as ‘literary’ in the Macmillan dictionary. In the utterance it is conventional but it also stands out and draws attention to itself as somehow different. As a result, listeners may pay attention to the contrast between its marked meta- phorical and its regular non-metaphorical meaning and wonder whether they are instructed to activate the basic sense of the term. When they do, they realize that the non-metaphorical sense is also relevant, since a political campaign does involve a lot of journeying. Elena Semino has called this topic-driven metaphor (Semino 2008). Because of its literary flavor, this turns into a form of wordplay, which cannot be anything but deliberate. Af- ter all, Obama could also have said ‘I want to thank my partner in this cam- paign’ and have conveyed the same point—but he did not. Deliberate metaphor hence requires some feature which alerts the ad- dressee that it is intended to be realized as a metaphor. Such a feature may come in the form of a lexical signal, as in simile, or in the form of additional relevant meaning, as with a breach of register that then results in wordplay. It may also come in the form of direct linguistic expression of the conceptual source domain which instructs the reader to set up a new, alien referential framework, as we saw in our systems development text. The most explicit, but also brilliantly frivolous, form of deliberate metaphor that I know of in this context is Shakespeare’s “Shall I compare thee to a summer’s day?” The function of these features of deliberate metaphor is to alert the ad- dressee to its intended communicative function. I suggest that this function is to change perspectives. When a metaphor is used deliberately, it instructs the addressee to momentarily adopt another standpoint, in another frame of ref- erence, and to reconsider the local topic from that point of view. In the jour- ney example, this rhetorical move yokes together the mental transformation caused by the political campaign with the physical effort required by its many journeys. The deliberate metaphor in a way doubles the sense of the appreciation that Obama extends to his Vice President-elect. In the systems development text, the doctor-patient domain offers a fresh and more familiar perspective on the complexities of the interaction between systems develop- ers and their customers, which is a much more restricted domain of experi- ence. Deliberate metaphor truly is a means of understanding one thing in terms of something else. Changing perspective is the main communicative motive for deliberate metaphor, which in turn may have various rhetorical goals (entertainment, information, persuasion, instruction, and so on). All of these ideas still need theoretical honing and development. One problem is the effect of a given deliberate metaphor on its environment. Thus, the journey metaphor does not change the non-deliberate status of the words in and this: it will be hard to imagine that addressees feel forced to set up some cross-domain mapping from space to processes for either of these 59 words. However, awareness of metaphor caused by the journey phrase may indeed have an effect on people’s experience of the conventional and indi- rect expression from the heart. The phrase seems to stand out from its con- text as well, simply because politicians are rarely associated with sincerity and its conventional symbol, the heart. Thus, even though in other contexts this phrase might be judged to be non-deliberate, in the present case I am not so sure. Attention to metaphor may be primed by previous awareness of metaphor. The theory of deliberate metaphor still needs more deliberation, discussion, and eventually, research. 5 Conclusion One of the important points about deliberate metaphor is that it concerns a communicative property. It has to do with a certain degree of awareness on the part of language users that they are using metaphor as a specific means of communication. It is in principle immaterial whether those metaphors are conceptually conventional or novel (even though it is clear that novel meta- phors will typically be deliberate, whereas conventional metaphors are not). And it is in principle immaterial, too, whether those metaphors are linguisti- cally indirect or direct (even though it is also clear that direct metaphors will typically be deliberate, whereas indirect metaphors are not). Both of these claims need to be expanded by the explication that this does not mean that all conventional and indirect metaphors are by definition non-deliberate, as we have seen. Another important aspect of the relation between deliberate metaphor and awareness has to do with processing. All deliberate metaphors by definition involve processing by comparison, or cross-domain mapping. This is be- cause they inevitably lead the addressee to some source domain from which they have to re-view the target domain. This is the essence of their commu- nicative function, which may be utilized for various rhetorical purposes. However, it does not follow that all non-deliberate metaphors are not pro- cessed by comparison. Non-deliberate metaphors may or may not be pro- cessed by comparison, but since this does not reach levels of consciousness, this is a question that cannot be answered by introspection and observation. Only experimental techniques seem to be able to tap these unconscious lev- els of metaphor processing. The issue of metaphor processing is currently the subject of a lively debate between two psycholinguistic schools of thought, which attempt to determine when metaphor is processed by compar- ison as opposed to categorization (Gentner & Bowdle 2008; Glucksberg 2008). The cognitive-linguistic assumption that all metaphor is processed by cross-domain mapping is a bold hypothesis, but the evidence that has so far been presented in its favor still needs to be interpreted with care (cf. 60 McGlone 2007). My own position is that it is more likely that there is a par- adox of metaphor, to the effect that most metaphor will turn out to be pro- cessed not by comparison but by categorization or, even more simply, lexical sense disambiguation (Steen 2008). The upshot of this three-dimensional approach is that we need to distin- guish between various linguistic forms of expression which are related to un- derlying cross-domain mappings of a conventional or novel conceptual struc- ture, which in turn can be used non-deliberately or deliberately by language users on specific occasions. When they use them deliberately, they draw their addressee’s attention to the source domain as an alien domain in the local dis- course, and instruct the addressee to consider some aspect of the target domain from that perspective. When addressees do not miss that type of instruction, they can be said to have some form of awareness of the source domain of the metaphor, and experience the metaphor as a deliberate rhetorical instrument. Whether they always have the cognitive opportunity and capacity to label this specific instrument as a metaphor is a separate issue. That would add a kind of meta-communicative awareness which would seal the process of deliberate metaphor recognition as a specific cognitive process. Let us finally return to the use of dream and alive at the beginning of Obama’s speech. Do they exhibit the necessary features that can alert the addressee to their use as metaphors? I have my doubts. The fact that dream harks back to Martin Luther King’s originally deliberate metaphor cannot be contested and is hard to miss; but this may only constitute an allusion and revive the original reference, not necessarily its metaphorical character. It is perfectly possible for a listener to process dream in association with the Dr King speech while staying within the target domain meaning of hopes. If this is the case, dream does not provide a trigger for the interpretation of alive as a deliberately used metaphor, either. But let us examine these and other cases by further research in which we collect some data from recipients and perhaps even from producers. For it is clear that there is much to be learned by further exploring this field of re- search. Since we now have a large store of reliably identified metaphorical expressions at our disposal, we can study their distribution and function, in order to design such experiments with a more natural flavor than sometimes takes place in present-day psycholinguistics. Author’s note I am grateful to the Netherlands Organization for Scientific Research, NWO, for Grant 277-30-001, “Metaphor in discourse: Linguistic forms, conceptual structures, cognitive representations”, which made it possible to do this re- search. 61 References Caballero Rodriguez, Rosario. 2006. Re-viewing Space: Figurative language in architects’ assessment of built space. Berlin: Mouton de Gruyter. Cameron, Lynne. 2008. ‘Metaphor and Talk’, in Gibbs, Raymond W. Jr. (ed.), The Cambridge Handbook of Metaphor and Thought. Cambridge: Cambridge Uni- versity Press, 197–211. —. 2003. Metaphor in Educational Discourse. London: Continuum. Charteris-Black, Jonathan. 2004. Corpus Approaches to Critical Metaphor Analysis. London: Palgrave MacMillan. Charteris-Black, Jonathan, & Andreas Musolff. 2003. ‘“Battered Hero” or “Innocent Victim”? A comparative study of metaphors for euro trading in British and German financial reporting’, in English for Specific Purposes 22: 153–176. Chiappe, Dan, John M. Kennedy, & Tim Smykowski. 2003. ‘Reversibility, Aptness, and the Conventionality of Metaphors and Similes’, in Metaphor and Symbol 18(2): 85–106. Clark, Herbert H. 1996. Using Language. Cambridge: Cambridge University Press. Crisp, Peter. 2008. ‘Between Extended Metaphor and Allegory: Is blending enough?’, in Language and Literature 17(4): 291–308. —. 2005. ‘Allegory, Blending, and Possible Situations’, in Metaphor and Symbol 20(2): 115–332. —. 2002. ‘Metaphorical propositions: A rationale’, in Language and Literature 11(1): 7–16. —. 2001. ‘Allegory: Conceptual metaphor in history’, in Language and Literature 10(1): 5–19. Crisp, Peter, John Heywood, & Gerard Steen. 2002. ‘Metaphor Identification and Analysis, Classification and Quantification’, in Language and Literature 11(1): 55–69. Croft, William & D. Alan Cruse. 2004. Cognitive Linguistics. Cambridge: Cam- bridge University Press. Gentner, Dedre & Brian Bowdle. 2008. ‘Metaphor as Structure-mapping’, in Gibbs (2008): 109–128. Gibbs, Raymond W., Jr. 2006. Embodiment and Cognitive Science. Cambridge: Cambridge University Press. —. 1994. The Poetics of Mind: Figurative thought, language, and understanding. Cambridge: Cambridge University Press. —. 1993. ‘Process and Products in Making Sense of Tropes’, in Ortony, Andrew (ed.), Metaphor and Thought. 2nd ed. Cambridge: Cambridge University Press, 252–276. Gibbs, Raymond W., Jr. (ed.). 2008. The Cambridge Handbook of Metaphor and Thought. Cambridge: Cambridge University Press. Giora, Rachel. 2008. ‘Is Metaphor Unique?’, in Gibbs (2008): 143–160. —. 2003. On Our Mind: Salience, context, and figurative language. Oxford: Oxford University Press. Glucksberg, Sam. 2008. ‘How Metaphors Create Categories—Quickly’, in Gibbs (2008): 67–83. —. 2001. Understanding Figurative Language: From metaphors to idioms. Oxford: Oxford University Press. Goatly, Andrew. 1997. The Language of Metaphors. London: Routledge. Goddard, Cliff. 2004. ‘The Ethnopragmatics and Semantics of “Active Metaphors”’, in Journal of Pragmatics 36(7): 1211–1230. 62 Heywood, John, Elena Semino, & Mick Short. 2002. ‘Linguistic Metaphor Identifi- cation in Two Extracts from Novels’, in Language and Literature 11(1): 35–54. Kintsch, Walter. 1998. Comprehension: A paradigm for cognition. Cambridge: Cambridge University Press. Kövecses, Zoltán. 2002. Metaphor: A practical introduction. Oxford: Oxford Uni- versity Press. Lakoff, George. 1986. ‘A Figure of Thought’, in Metaphor and Symbolic Activity 1(3): 215–225. Lakoff, George & Mark Johnson. 1999. Philosophy in the Flesh: The embodied mind and its challenge to western thought. New York: Basic Books. —. 1980. Metaphors We Live By. Chicago: Chicago University Press. McGlone, Matthew S. 2007. ‘What is the Explanatory Value of a Conceptual Meta- phor?’, in Language & Communication 27: 109–126. Pragglejaz Group. 2007. ‘MIP: A method for identifying metaphorically used words in discourse’, in Metaphor and Symbol 22(1): 1–39. Rundell, Michael (ed.). 2002. Macmillan English Dictionary for Advanced Learners. Oxford: Macmillan Publishers. Semino, Elena. 2008. Metaphor in Discourse. Cambridge: Cambridge University Press. Shen, Yeshayahu & Noga Balaban. 1999. ‘Metaphorical (In)coherence in Dis- course’, in Discourse Processes 28(2): 139–154. Steen, Gerard J. 2008. ‘The Paradox of Metaphor: Why we need a three-dimensional model of metaphor’, in Metaphor and Symbol 23(4): 213–241. —. 2007. Finding Metaphor in Grammar and Usage: A methodological analysis of theory and research. Amsterdam: John Benjamins. —. 2004. ‘Can Discourse Properties of Metaphor Affect Metaphor Recognition?’, in Journal of Pragmatics 36(7): 1295–1313. —. 2002a. ‘Metaphor in Bob Dylan’s “Hurricane”: Genre, language, and style’, in Semino, Elena & Jonathan Culpepper (eds), Cognitive Stylistics: Language and cognition in text analysis. Amsterdam: John Benjamins, 183–210. —. 2002b. ‘Towards a Procedure for Metaphor Identification’, in Language and Literature 11(1): 17–33. Steen, Gerard J., Ewa Biernacka, Aletta G. Dorst, Anna A. Kaal, Irene López- Rodríguez, & Trijntje Pasma. 2010. ‘Pragglejaz in Practice: Finding metaphori- cally used words in natural discourse’, in Low, Graham, Lynne Cameron, Alice Deignan & Zazie Todd (eds), Researching and Applying Metaphor in the Real World. Amsterdam & Philadelphia: John Benjamins. Steen, Gerard J., Aletta G. Dorst, J. Berenike Herrmann, Anna A. Kaal, Tina Krennmayr, & Trijntje Pasma. 2010. A Method for Linguistic Metaphor Identifi- cation: From MIP to MIPVU. Amsterdam: John Benjamins. Sweetser, Eve. 1990. From Etymology to Pragmatics: Metaphorical and cultural aspects of semantic structure. Cambridge: Cambridge University Press. Van Dijk, Teun A. & Walter Kintsch. 1983. Strategies of Discourse Comprehension. New York: Academic Press. 63 Session papers Ludmilla A’Beckett ACADEMIC LIFE IS WAR: The Extended Metaphor in Tom Sharpe’s Porterhouse Blue Abstract. The distribution of WAR metaphors in Tom Sharpe’s Porterhouse Blue is the focus of this paper. A technique for cataloguing instances of humour (Attardo 2001, 2002) has been used for finding various semantic links between metaphors in text. Groups of metaphors are analysed as indicators of the mind style of protago- nists (Fowler 1996 and Semino 2002) and plot dispersion (Scott & Tribble 2006). Different approaches for grouping metaphors within a text and the totality of texts, e.g. discourse, are discussed. Implications and applications of clustering metaphors in fictional texts are suggested. Keywords: extended metaphor, characterisation, mind style, plot dispersion, meta- phor clusters 1 Introduction In this paper, the role of WAR metaphors in the representation of the conflict and characterisation of the main protagonists of Tom Sharpe’s Porterhouse Blue has been analysed. I have followed a technique suggested by Attardo (2002) for analysis of humorous instances in Oscar Wilde’s Lord Arthur Savile’s Crime. Attardo demonstrated the distribution of humorous effects around the verbal fabric of Lord Arthur Savile’s Crime and drew conclusions about the role of these effects in producing the global coherence of the nov- el. Similarly, I will examine some strings of metaphors, comment on their structure and the mental models they highlight. Although this paper remains within the tenets of conceptual metaphor theory (Lakoff & Johnson 1980), it reveals some tensions between samples of conceptual metaphors, e.g. ARGUMENT IS WAR and the metaphor systems used in the text. Even though clusters of metaphors in the novel follow some conventional mappings, e.g. between ARGUMENT and WAR, they introduce the idiosyncratic experience of the protagonists. The tradition of linking patterns of metaphorisation to goals, motifs, be- liefs, traits, emotions and the social relationships of fictional characters has been developed within the approach of cognitive stylistics (Semino 2002, Culpeper 2001, 2002, Hamilton 2002, Popova 2002, Steen 2002 and others). 67 Cognitive stylistics is a rapidly expanding field at the interface between lin- guistics, literary studies and cognitive science (Semino & Culpeper 2002a: IX). Studies in cognitive stylistics often focus on the role of language ex- pressions in construing more abstract non-literal concepts or frames of knowledge. Following established tradition, this paper singles out a few themes reinforced through the use of metaphoric language. Metaphor pat- terns point to salient behavioural patterns and reveal the inevitability of the tragic end in the novel. The notion of ‘mind style’, introduced by Fowler (1996) and elaborated in Semino (2002), has been used to define the main area of metaphor projection. Fowler (1996: 214) suggests that mind style is “the world view of an author, or a narrator, or a character, constituted by the ideational structure of the text”. Semino (2002: 99) comments that “the no- tion is most useful where narratives involve the foregrounding of linguistic patterns that suggest some salient cognitive habit or deficit”. The mind style of characters may consist of a blend of global trends with the specific cir- cumstances in which protagonists find themselves, e.g. a conceptualisation of aggressive behaviour in terms of war attenuates the particulars of the aca- demic environment. In sum, metaphors have been considered in this paper as devices channelling readers’ attention to the culminations in the plot and the characterisation. 2 Plot overview In Porterhouse Blue, Tom Sharpe describes a confrontation between the Master of a Cambridge College and the Senior Fellows over the school poli- cy-making. The main conflict develops throughout discussions during meet- ings of the Council. The subject matter of the discussions varies. The topics include enrolment procedures, the raising of money for college restoration, the gastronomic preferences of the Fellows, the instalment of a condom dis- penser and other issues. Only the intentions of the participants remained constant. All sides of the conflict seek unlimited power over College matters and humiliation of their opponents. The Master, Sir Godber, is a lone fighter supported by his wife Lady Mary and the Bursar, who defects from the Fel- lows’ camp. The Master’s main antagonist is the Dean, who is reconciled with the Senior Tutor at the prospect of a major challenge to their group power. The Dean is a vitriolic man and notorious for his spiteful comments and provocations. The Senior Tutor abandons his scholarly ambitions for the sake of college sports achievements and gastronomic pleasures. Both men are feared because of their outbursts of fury. The resistance also includes the Praelector and the deaf Chaplain. Despite the weight of numbers, Sir Godber is invincible in debates. The tragic death of one of the students and damage to the college properties are 68 the only impediments to his victorious march to absolute control of college matters. His victories are assured by his previous experience in politics. Be- fore debates he carefully analyses the situation and identifies the weaknesses of his opponents. Then he comes up with a suggestion which seems radical and enraging at the time. While his furious enemies are engaged in attacking his initial suggestion, he would shift ground and demand more concessions. His first proposal would look innocent compared to subsequent demands. The Fellows’ rage precludes them from a thorough assessment of his pro- posals. They are driven by the necessity to agree upon something to avoid accusations of being dismissive and obstructive. The Master uses the perpet- ual disagreements within the group of his opponents to his own advantage. Sir Godber can hardly be perceived as a real reformer either during his ca- reer in politics or in his retirement as Master of the Cambridge college. His slogans “Every house to have a bathroom” (p. 3) or “Alterations without changes” (p. 30) are indicative of the degree of novelty he tries to imple- ment. However, in a Porterhouse still shrouded in a medieval atmosphere and with their adherence to the medieval code, any change was viewed as a threat and a revolution. Devastated by the Master’s defiance and their own incapacity to run de- bates, the Fellows resort to blackmail with assistance of their mighty friends. They appeal to their former graduates General Cathcart, who has friends in high places, and the journalist Cornelius Carrington to spin bad publicity around the Master. Sir Godber would undoubtedly have used their tricks to his own advantage but for a dark horse, the Senior Porter in Porterhouse, Skullion, who is sacked for his arrogance and idleness. Skullion is the em- bodiment of all the college traditions and gains the sympathy of the British public by revealing the inhuman nature of the changes introduced by the Master and the wrongdoings of the Fellows. Skullion’s revelations are a serious blow to Sir Godber’s ambitious intentions. After regaining his posi- tion of control and another victory over the Fellows, Sir Godber is ap- proached by Skullion, who wishes to get back his position as College Porter. During this unexpected meeting, the refused and offended Skullion mortally wounds the Master by accident. Sir Godber dies in the arms of his enemies, the Fellows, naming Skullion as his ‘assassin’. However, the Fellows read the name as the Master’s last wish to nominate his successor as the master of the college. Skullion, tormented by guilt and fear, suffers a stroke. Having a disabled man as their Master suits the Senior Fellows well, who can again enjoy their total control over policy-making in Porterhouse. 69 3 Distribution of metaphors in the novel WAR metaphors are salient in the novel, as both the Master and the Fellows conceptualise their relations in military terms. By ‘metaphors’ I understand “various surface language forms, such as similes, analogues and hyperbole, but only those instances which are metaphorical, in the sense of bringing two disjunctive domains together” (Cameron & Stelma 2004: 117). Sir Godber and the Fellows constantly re-assess their positions using expressions from the domain of WAR. However, the interrelated and adjacent metaphors may originate from dif- ferent source domains. The words’ etymology and context cues can point to different thematic fields. Given the variety of vehicles that manifest multiple facets of the conflict, it would seem appropriate to discuss the conceptual dimensions they represent. Since the vehicles are thematically connected, it is also important to provide some ideas on distance within each group of interrelated metaphors. Military metaphors, e.g. war, to surrender, to attack, weapons appear fre- quently. However, various degrees of metaphoricity may be observed among these so-called military terms. On the one hand, the degree of metaphoricity shows whether the metaphor vehicle is conventionally used for naming the given referent or topic (Deignan 2005: 37–47). On the other hand, the degree of metaphoricity suggests a perception of the distance between the source and target domains (Cameron 1999a). For instance, the words tri- umph/triumphant are conventionally used for description of positive feelings associated with achievements, regardless of the nature of the achievements. Triumphant is used to highlight any success and happiness upon completing a task, though etymologically it relates to military victory. Triumphs or mili- tary parades displaying attributes of victory can hardly be observed in day- to-day routines and therefore the word triumph seems to be weakening its links with its military prototype and broadening the scope of its application. Similarly, victory and to win may not always require transposition of proper- ties from WAR domains to domains of personal achievement, e.g. sports, music, completion of an assignment, and others. They refer to the situation of being the first in a competition or completion of a task, regardless of the circumstances of this achievement. However, if the context supports the idea of ongoing fights with enemies, the transposition from the WAR domain is more evident. The number of transferred properties increases, e.g. humilia- tion of enemies, display of victorious attributes, imposition of tribute and the victor’s ability to make rules. Therefore, some words like belligerent, victo- ry, triumph acquire a “higher metaphorical status” (Cameron 1999a: 107) within the context of constant battles among the academy administrators. Some metaphors point to aggressive behaviour or confrontations but have not originated in the WAR domain. Other source domains that support the 70 idea of violence and create coherent patterns with WAR metaphors include WEATHER, HUNT, IRON and ANIMAL BEHAVIOUR. For instance, in (1), anger is represented as a storm and as an action of God. In (2), stormed is linked to the idea of unpredictable changes in the real world. In both examples, storm signals a confrontation and danger. (1) The Tutor had come storming out of Chapel with his gown billowing behind him like the wrath of God. (p. 88) (2) A lifetime of little duties easily attended to while the world outside stormed by in a maelstrom of change had bred in Skullion a devotion to the changelessness of Porterhouse tradition. (p. 19) The domain of ANIMAL BEHAVIOUR has supplied concepts for understanding ANGER, INSTINCTIVE REACTIONS and RETREAT. (3) What sweet maids required was a first-rate education and what sleep- ing dogs [=the Dons —LA] needed was a kick up the backside. (p. 31) (4) The Dean was barking up the wrong tree after all. (p. 109) (5) ‘Damn the Master,’ the Dean snarled, struggling to his feet. (p. 106) (6) [The Fellows] crawled to him and Sir Godber indulged himself in the recollection before going on to consider the implication of their sur- render. (p. 117) (7) [T]here had been no need for the Dean, of all people, to wag his tail so obsequiously. (p. 117) Along with weapons, guns and shrapnel, HUNTING and FISHING devices, e.g. baits and traps, have been used in Porterhouse ‘battles’ with opponents. Metaphors from source domains other than WAR often create consistent clusters within the megagroup representing the confrontation of the Master with the Fellows. Therefore, thematic subclasses of the megagroup are in- formative of the mode of conduct adopted by protagonists at a certain point. Metaphor vehicles display structural variation, e.g. they can consist of a single unit and ‘chunks’. “‘Chunking’ language means recognising a group of two or more words that regularly co-occur” (Deignan 2005: 121). Chunks are usually represented by idioms, e.g. barking up the wrong tree in (4), to wag one’s tail in (7) and others. “Idioms are indivisible units whose compo- nents cannot be varied or varied only within definable limits” (Deignan 2005: 195). In Porterhouse Blue and other Tom Sharpe novels, they intro- duce multiple meanings and can often be broken down into components. On the one hand, the Dean behaved like a dog and on the other, he made wrong assumptions in (4) and behaved submissively in (7). The conventional fig- urative meaning of idioms can be accessed together with their prototypical meanings and connotations, e.g. dogs’ reflexes and their associated negativi- 71 ty. Puns have been created, since multiple senses have been accessed. Leech (1969: 209) defines a pun as a foregrounded lexical ambiguity. Metaphors of aggression and reactions to aggressive behaviour constitute a loose overarching theme. Within this general network of metaphors differ- ent subdomains emerge. Subgroups of metaphors can be traced either through specification of their area of projection or subdivisions of source concepts. Configurations of the source domain have been previously shown within the subgroup ANIMAL BEHAVIOUR. The target domains of the meta- phors may have a number of dimensions, too. On a global level, their projec- tion reveals the particulars in the conflict between the Master and the Fel- lows. However, some segments target the Dons’ beliefs and attitudes, while others are applicable to the characterisation of the Master. Some subgroups represent meetings of the Council and conversations, while others reveal expectations of the protagonists. A few oppositions can be located through the analysis of metaphor applications: 1. The Dean’s perception of the situation versus the Master’s percep- tion; 2. The Dons’ actions versus the Master’s actions; 3. The perception of the situation through the eyes of the main oppo- nents (The Master and the Dons) and of the supporters (Lady Mary, Skullion, General Cathcart, and Cornelius Carrington); 4. Real events and evaluative responses to them, e.g. meetings and the TV show, versus anticipations of changes, e.g. plans, and fears; 5. Verbal confrontation, e.g. insults, arguments and speeches, versus actions, e.g. dismissals, implementation of policy and other activi- ties. The general network of military metaphors and adjacent metaphors of ag- gression consists of loosely related groups overlapping in parts and creating numerous subsystems of multidimensional mappings and perspectives of reality (drawn from the description of metaphor networks in Deignan 2005). Metaphors of aggression constitute a range of experiences that can be recre- ated on various level of generalisation, e.g. DOG BEHAVIOUR versus ANIMAL BEHAVIOUR. The network encompasses the ARGUMENT IS WAR conventional system but also includes other idiosyncratic mappings, e.g. many vehicles introduce AGGRESSIVE INTENTIONS and BEHAVIOUR outside of the domain of verbal confrontation. The metaphors produce high ‘multivalency’ (Goatly 1977: 258–259) or a very wide ‘scope’ (Kövecses 2000, see also Semino 2002: 112). The distance between occurrences of metaphor vehicles contrib- utes to difficulties in recreating the network. There are instances where a few metaphors occur within the same proposition, e.g. to bait and the trap (p. 16). There are segments of high density within a single paragraph. There are many metaphors which can be linked across pages and chapters, e.g. the DOG 72 metaphor (pp. 31, 106, 109, 117). Therefore, my reproduction of metaphor patterns aims at the demonstration of multiple flexible connections between vehicles. 4 Approaches to grouping metaphors A number of terms and frameworks exist for the representation of metaphor groups in naturally occurring language data. Adoption of any one of them would lead to a certain emphasis, with the possible exclusion of some as- pects in research. Therefore, the scope for each framework has been revised. In this section, I provide an overview of the grouping technique represented through the following terms: metaphor clusters or bursts (Cameron & Stelma 2004), extended metaphor or megametaphor (Kövecses 2002), mininarrative or metaphoric scenario (Musolff 2006) and metaphor theme (Black 1993). Cameron & Stelma (2004), investigating reconciliation talks between a perpetrator of violence and a victim, found that speakers produced a range of metaphorically used words and phrases when some intensive interactional work linked to the overall purpose of the discourse had been carried out. They presented a visualisation methodology for identifying and exploring metaphor clusters. The areas of metaphor density revealed their relevance both to a particular episode in the talk and to the overall dynamics of dis- course. Further investigations of the metaphor clusters (Cameron & Low 1999a, Littlemore & Low 2006) found that many clusters contain peripheral or non-core expressions. They are called ‘outliers’, e.g. the words that reveal a greater metaphoric intensity when they occur in the company of a clearly metaphoric term (Littlemore & Low 2006: 138). The status of triumph, vio- lence (as a verbal behaviour) and victory (see Section 3) can be explained through the phenomenon of outliers. Clusters may accommodate more than one perspective and more than one extended metaphor (Cameron & Stelma 2004: 132). Kövecses (2002: 53) defines the extended metaphor or megametaphor as a group of micrometaphors that are coherent. However, the coherence of micrometaphors can often be established both on the local level which repre- sents a particular episode or a particular viewpoint and the global level that blends all episodes and characters into a megasituation (see Alm-Arvius 2008 on cohesive functions of tropes). From a global perspective, the aca- demic life in Porterhouse Blue is consistently conceptualised in terms of WAR or enduring violence. Clusters or local systems of metaphors (drawn from Cameron 1999b: 16) support the megametaphor ACADEMIC LIFE IS WAR but provide a specification from different perspectives. Various mi- croscale mappings derive from the main domains of the megametaphor. My understanding of the megametaphor is similar to the definition of metaphor- 73 theme in Black (1993: 24), which is an abstraction from verbal manifesta- tions of the metaphor and contexts in which it does or might occur: “Meta- phor-theme is available for repeated use, adaptation, and modification by a variety of speakers or thinkers on any number of specific occasions”. A few speakers or characters in Porterhouse Blue make their contribution to the megametaphor. The megametaphor constituents can be grouped ac- cording to a perspective taken within the global system, e.g. protagonists’ mind style or other target domains invoked. There are a number of ways for establishing connections between minimetaphors (Kövecses 2002) or meta- phor ingredients (Black 1993). One of the ways suggested by Cameron & Stelma (2004) is to look at local co-occurrences of metaphors and identify boundaries of clusters within the textual fragment. Another way is to group vehicles from the same source domain across the whole book or establish systematicity in the metaphor projection. Musolff 2006 adopted the strategy of finding sub-divisions within the source domain for profiling metaphor themes in the EU discourse. A specific configuration of source concepts established through connections among vehicles has been named a metaphor scenario or mininarrative (Musolff 2006). For instance, within the FAMILY source domains, MARRIAGE and PARENTHOOD scenarios have been found. Within the WAR source domain, subsystems such as FRAY or COMBAT and others can be profiled. In other words, syntagmatic and paradigmatic con- nections emerged as a result of the megametaphor analysis. 5 Method of modelling metaphor connections in the novel The principles of cataloguing tropes suggested by Attardo (2002) have been adopted for representation of the megametaphor structure. The application of his technique (summarising Attardo 2002: 234–240) requires a) analysis of the text as a vector, where metaphors are identified as a deliberate cross- domain mapping (Steen 2007, 2008); b) grouping metaphors that co-occur (the local level) or are thematically linked on the global level. Attardo’s principles of trope extraction covered instances of humour iden- tified as per the general theory of verbal humour. The transposition of his technique for the purposes of this investigation involves classification of connections between metaphor units in text. Two types of relations have been envisaged: a) A group of metaphors that are locally connected and related to the same segment of text is called a strand. The analogy can be drawn between the strand and a metaphor cluster (Cameron & Stelma 2004). A minor variance from the notion of cluster relates to fuzzy boundaries of the strand: the same episode description may run over 74 a dozen pages. The clusters in Cameron & Stelma (2004) seem to have compact boundaries and are distributed across adjacent sen- tences. b) A strand of strands is called a stack (Attardo 2002: 236). For in- stance, the metaphor group used for the description of the first meet- ing of the Council can be linked to other events in the Council. The target domain of this stack can be presented as MEETINGS OF COUN- CIL. Sometimes the source concepts of metaphors point to a subfield within the general domain, e.g. within the WAR domain, the groups of WEAPONS or STRATEGIC THINKING can be identified. Such groups of metaphors are simi- lar to the mininarrative as defined by Musolff (2006). They can be covered by the notion ‘stack’ but with some conditions: mininarratives are metaphor units selected from various strands across the novel rather than a coalescence of adjacent strands (see the THE DEAN IS A VICIOUS DOG metaphor in section 4). Therefore, megagroups of metaphors can be compiled following affinities within the source or target domains. For instance, it is possible to compile a stack on the DEAN mind style or the MASTER mind style, collecting meta- phors from the same target domain or area of projection of metaphors. On the other hand, the stacks ANIMAL BEHAVIOUR or HUNTING can be compiled by following the subdivisions within the source domain ANIMALS. All identified strands have been linked to episodes and assigned a label that would be indicative of a particular segment in the plot. Some segments are represented by a few metaphor clusters or stacks relating to the same situation. The thorough quantitative analysis constitutes another aspect of the research and therefore, for the purpose of this paper, it has been omitted. It is debatable whether a composite metaphor represents a single unit or a number of constituents, e.g. to win the war and what sleeping dogs needed was a kick up the backside. Attardo (2002) and Cameron & Stelma (2004) presented their findings in graphs. I have refrained from any visual representation of trope density, as there are numerous problems with the countability of words and metaphor units in segments. It would be well nigh impossible to represent all the metaphors in the novel, and therefore I only focused on those that, in my opinion, were used deliberately (Steen 2008) for the purpose of foregrounding (Alm-Arvius 2008) the ongoing confrontation between the Master and the Fellows. 75 6 Results If patterns of metaphoricity are profiled against the timeline of events rather than the characters’ mode of conduct, then the following segments attract the largest metaphoric representations: 1. Attitudes and plans before the first meeting of the College Council 2. Outcomes of the first meeting of the Council 3. Attitudes and plans before the second meeting of the Council, ad- justed for the tragic death of a student and destruction of College property 4. The second meeting of the Council 5. Attitudes and plans for the TV programme on Porterhouse with the journalist Cornelius Carrington 6. The third meeting of the Council 7. Outcomes of the third meeting of the Council The weight of each segment in terms of the number of metaphor units and perspectives presented varies. The first segment, ‘Attitudes and plans before the first meeting of the Council,’ consists of two stacks: ‘The Master’s plans and attitudes’ and ‘The Fellows’ plans and attitudes’. The first stack, on the Master’s beliefs and actions within the first segment, consists of five strands whose length varies between a page and a paragraph. The Fellows’ attitudes and beliefs within the first segment consist of three strands. The Master’s characterisation thus attracts more metaphors in comparison with the Fel- lows within the first segment. The strands on the Master’s beliefs are intro- duced either by the narrator or the Master himself. The strands on the Fel- lows are introduced by the narrator, the Fellows (chiefly by the Dean and the Senior Tutor) and Skullion. The second segment, ‘Outcomes of the first meeting of the Council,’ con- sists of three strands, or one strand on the Master’s assessment of the situa- tion and one stack on the Fellows’ assessment of the situation. The stack of the Fellows’ assessment of the situation presents a strand on the Dean’s atti- tudes and plans and the Fellows’ mood. The composition of this segment is straightforward and therefore can be reproduced for the purpose of illustra- tion. (8) SEGMENT 2. STRAND 1. The Dean’s and the Fellows’ perspectives on the out- come of the first meeting of the Council. ‘Damn the Master,’ the Dean snarled, struggling to his feet. ‘I said he must be stopped. I didn’t say we had to surrender to the swine’. He 76 waddled to the head of the table, portly, belligerent and stubborn, like some crimson toad and with all that creature’s resilience to the chal- lenges of climate. (p. 106) [W]e have powerful friends. [W]e have powerful allies. (p. 107) ‘I am only concerned with results. Humble pie, you said yourself. Very well, if Sir Godber requires humble pie to retract his threat he shall have it. I shall see to it that he eats it himself later. Besides I should not like him to think we are in any way divid- ed.’ He stared fiercely at the Bursar. ‘At the time of crises it is vital that we present a united front.’ (p. 108) STRAND 2. The Master’s perspectives on the outcome. They trooped out and Sir Godber, having seen them to the door, stood at the window watching the dark procession disappear into the winter evening with a new sense of satisfaction. ‘The iron fist in the iron glove,’ he murmured to himself, conscious that for the first time in a long career of political manoeuvring and compromise he had at long last achieved a clear-cut victory over an apparently intransient opposition. They crawled to him and Sir Godber indulged himself in the recollection before going on to consider the implication of their surrender. No one—and who should know better than Sir Godber— crawled so submissively without good reasons. [T]here had been no need for the Dean, of all people, to wag his tail so obsequiously. ‘Buy- ing time,’ he thought shrewdly, ‘but time for what?’ (p. 117) STRAND 3. The Dean’s and the Felloаs’ mood after the first meeting of the Council. At High Table the Fellows dined in moody silence. Even the Chef’s poached salmon failed to raise their spirit, dampened by the obduracy of the Master and the memory of their capitulation. Only the Dean remained undaunted, shovelling food into his mouth as if to fuel his determination and mouthing imprecations on Sir Godber simultane- ously. (p. 118) The third segment, ‘Attitudes and plans before the second meeting of the Council, adjusted for the tragic death of a student and destruction of the Col- lege property’, could be treated as a part and as an extension of the second segment. It has been organised into an independent strand because there is a narrative disruption between the second segment (‘Outcomes’) and the third segment (‘Attitudes and plans... for the second meeting’). A peripheral plot line (Attardo 2002:241) on the attraction between Zipser, the graduate stu- dent, and Mrs. Biggs, his bedder, breaks the main plot. The peripheral seg- ment has not been included in my analysis but the plans of the conflicting parties undergo considerable change as a result of the tragic act. Therefore, the plans and attitudes generated in response to the peripheral plot develop- 77 ment are considered as an independent segment. Again, given the simplicity of the segment structure, it (or at least its major units) can be reproduced for illustrative purposes. (9) SEGMENT 3. STRAND 1. Ladв Marв’s perspectives on the forthcoming meeting of the Council. ‘The Dean,’ said Lady Mary with feeling, ‘is a male chauvinist pig. A sensible policy of coeducation would avoid the sexual repressions that result in fetishism. You must make this point at the next Council meeting...’ Lady Mary regarded him sternly. ‘Godber’, she said, ‘you must not weaken now. You must not compromise your principles. You must stick to your guns.’ ‘Guns, my dear?’ ‘Guns, Godber, guns.’ ... ‘Godber, you are prevaricating,’ said Lady Mary firmly. (pp. 155–157) STRAND 2. The Master’s initial stance toward the forthcoming meet- ing of the Council. ‘It’s going to be very difficult to quell the Dean now,’ he said thoughtfully. Sir Godber raised his eyebrows doubtfully. What guns he had had, and, in the light of Lady Mary’s pacifism, he doubted if the metaphor was morally appropriate, appeared to have been effec- tively spiked by Zipser’s tragic act. (pp. 154–155) STRAND 3. The Master’s plans adjusted. He took his coffee cup and went through to his study and sat mo- rosely by the fire and wondered at his own pusillanimity. ‘We didn’t win the war,’ thought Sir Godber, ‘we just refused to lose it.’ Stirred to a new belligerency, he reached for the poker and poked the fire an- grily. (p. 157) Sir Godber continued, ‘I am afraid it will be an uphill battle.’ (p. 160) Segment 4 is a concise summary of the third meeting of the Council. The metaphorical representation of the event runs across four pages at least. However, the representation of the meeting adheres to the unity of locus, time and action. The segments consists of three strands which mirror Skul- lion’s, the Dean’s and the Master’s perceptions. Again, this segment can be easily reproduced for illustration of the structure of the segment or stack. 78 (10) SEGMENT 4. STRAND 1. Skullion’s monitoring. In the boiler-room Skullion followed the argument with difficulty. The Master’s tactics evaded him. [...]A new fury possessed Skullion, a bitterness against Sir Godber that was no longer a concern for the tra- ditions of the College...but a sense of personal betrayal. Above his head a new violence of debate had broken out in the Council Cham- ber. Sir Godber had announced the proposed installation of a contra- ceptive dispenser. (pp. 170–172) STRAND 2. The Dean’s reactions to the meeting of the Council. The Dean erupted from the meeting with a virulence that stemmed from the knowledge that he had been outmanoeuvred. The Master’s appeal to principle had placed him in a false position and the Dean was conscious that his argument against the Master’s proposed econ- omies had lacked the force of conviction. The Bursar’s sudden change of allegiance had infuriated him too.[...]The Dean cursed the Bursar viciously as he climbed the stairs to his room. There remained only Sir Cathcart and already he had shown himself pusillanimous in the mat- ter of calling a meeting of the Porterhouse Society. (p. 172–173) STRAND 3. The Master’s attitude and aspirations. With the silent thought that it was time such relics of the past [chief- ly about Skullion —LA] got their marching orders, the Master led the way into the Lodge. (p. 174) There are four events which represent ‘battles’ between the Master and the Fellows, e.g., an open confrontation and an exchange of arguments and un- pleasantness. The direct projections of the ARGUMENT IS WAR system are three meetings of the Council and the TV programme with Cornelius Car- rington. The second meeting of Council is the only scene of the war that has received a broad metaphor description from multiple perspectives. The first meeting of the Council has been defined through a single metaphor high- lighting the Fellows’ strategy and attitudes. (11) The members of the Council stared at [Sir Godber] with open bellig- erence. (p. 102) Similarly, the discussion and views given in Cornelius Carrington’s pro- gramme on Porterhouse attracted the key metaphor ‘crusade’ (example (12)). The ‘battles’ are represented by fewer metaphors than the plans, attitudes, retrospective analysis and emotional responses. 79 (12) By the time the programme ended, the switchboard at the BBC was jammed with calls from people all over the country supporting Skul- lion in his crusade against the present. (p. 271) Segment 5, on preparations for the TV programme on Porterhouse, is the largest area of metaphor use. The segment consists of strands representing beliefs, plans and moods of the Fellows and the Dean in particular. It seems that after the second meeting of the Council the Fellows took the initiative in hand. Therefore, the whole segment has been packed with metaphors ad- dressing the mind style of the Dons. It includes two strands describing the explosive atmosphere in the Dining Hall, and one strand on a conversation of the Dean with General Cathcart, in which the General’s stock of metaphors inspired the Dean. The Dean’s plans for and attitudes toward the college form a distinctive strand. The segment also includes the set of beliefs of Cornelius Carrington and the set of beliefs of Skullion. Both characters have been used as a weapon in fighting with the Master. The conversation of the Dean with Cornelius Carrington encompasses a strand of metaphors that reflect the process of destruction of the journalist’s self-respect. The total number of strands in Segment 5 is ten if Lady Mary’s view in the conversa- tion with the Bursar and zoological metaphors on Dons as prehistoric ani- mals are taken into account. The clear borderline between Segments 5 and 6 is the metaphor on Skullion’s input into the academic animosity, represented in example (12), above. Segment 6 focuses on the last meeting of the Council and its preparations. Two strands from this section represent a) the preparations of the Dean and the Senior Tutor (pp. 286–287) and b) the situation in the Council (pp. 288, 291, 292). The latter strand consists of the metaphor units addressing the Senior Tutor’s mode of conduct, e.g. ‘he led the attack’ (p. 288), the Mas- ter’s reaction, e.g. ‘the Master snapped’ (p. 288) and the Master’s characteri- sation of the Fellows, e.g. ‘custodians’ (p. 291). It concludes with the meta- phor on the Master’s state of mind, e.g. ‘emotionally depleted’ (p. 292). It is a relatively small strand which runs, nevertheless, over almost six pages. The last segment, ‘Outcomes of the third meeting of Council,’ consists of four strands. Two strands represent evaluations of the situation a) in the eyes of the Dons and b) in the eyes of the Senior Tutor. The two strands or one stack include synonymous metaphors, e.g. ‘to succumb at last to the least effectual of the politicians’ versus ‘the Master’s victory’ and ‘the guardians of the political inertia’ versus ‘safeguards against the terrors of thought’ (p. 293). In the second strand the Senior Tutor suggests ‘murder’ as a hyperbole for dealing with the Master. However, the joke would become a prophecy later. The third strand contains Lady Mary’s view on the conflict and the Master’s view on the conflict. It is an assessment of the third meeting of the Council which has evolved under Lady Mary’s influence and choice of met- aphors. The last strand represents the blackmailing of General Cathcart and 80 the Government by the Dean and the Senior Tutor. The murder of the Master by Skullion ended the WAR of arguments. The source and target domains of the metaphor ARGUMENT IS WAR are reversed at the end of the novel. The argument has ended up as the physical elimination of the enemy rather than destruction of his arguments. My identification of segments, strands and stacks of metaphors has re- quired consideration of a) the length of each text fragment or segment; b) the relevance of metaphors for the development of the plot, e.g. the conflict be- tween the Master and the Fellows; c) variation in time, location and charac- ters involved in each segment; d) affinities in the area of projection of meta- phors and their source domains. Some metaphors may look dead, e.g. to succumb, pusillanimity, allegiance and to prevaricate. However, within the context of ongoing battles, the metaphors are revived and contribute to the impression of confrontations between two armies rather than individuals. The distribution of metaphors across segments of the novel reveals initia- tives, intensity of animosity and positions of power. 7 Discussion 7.1 Functions of metaphor groups Prior to the commencement of the analysis, I had some expectations of find- ing correlations between metaphor groups and characterisation of the parties with the conflict and intensity of animosity. The results have exceeded my best anticipations. The distribution of strands relating to the plans of the Master and his allies and those of the Fellows and their allies reveals at- tempts undertaken and initiatives involved. If, in the segments on prepara- tions for the first and second meeting of the Council, the Master takes the lead, the Fellows increase the number of their efforts in preparation for the last meeting and the programme with Cornelius Carrington. As the Fellows were the party which was always defeated in open confrontation, the number of strands devoted to their reflection upon events in the Council is larger than those of the Master. There are also interesting strands and stacks repre- senting the input of allies to the battles: the stock of metaphors used by Gen- eral Cathcart was transferred to the Dean; the stock of Lady Mary’s meta- phors was infused into the Master’s stock. Popova (2002: 50) and Alm-Arvius (2008) argue that figurative language often lends coherence to various elements of text structure. According to Popova (2002: 66), systems of metaphors in a novel act as thematic structur- ing devices operating on two levels. On the level of comprehension or short- term memory they govern our understanding of the textual information. On the level of interpretation or long-term memory the metaphor systems chan- 81 nel our conscious reflection. The analysis of metaphor groups in Porterhouse Blue corroborates these observations. Metaphors trigger the reader’s judge- ment about the characters and the nature of their debates. The role of meta- phors in structuring the textual flow is evident. The patterns of dense meta- phoricity represent the plans and attitudes of the enemies. Key metaphors demonstrate the essence of their discussions, e.g. belligerence, tactics, be- trayal, outmanoeuvred, erupted, crusade, attack. The metaphor scenario ARGUMENT IS WAR frames the structure of the novel. 7.2 Metaphor scenarios and the Master’s mind style More specific examples can be provided in relation to the effect of meta- phors on shaping judgement about a character. If we focus on the notion of mininarrative or metaphor scenario introduced by Musolff 2006, it is possi- ble to see characters’ habits in more details. The metaphor scenario is a met- aphor theme or a configuration of source concepts sifted from a narrative. Multiple metaphors used for the characterisation of the Master can be grouped under the label STRATEGIC THINKING. (13) Stack STRATEGIC THINKING This was precisely what he intended to administer to Porterhouse. Sir Godber planned his campaign. (p. 31) It was precisely on such divisions of opinion that he thrived. (p. 32) Satisfied with his plan of campaign, the Master turned on his side and fell asleep. (p. 33) The atmosphere was just what he wanted for the announcement of his plans. They would react predictably and with a violence that would disarm them. (p. 98) ‘The iron fist in the iron glove,’ he murmured to himself, conscious that for the first time in a long career of political manoeuvring and compromise he had at long last achieved a clear-cut victory over an apparently intransient opposition. (p. 117) The Master’s tactics evaded him. (p. 170) The Dean erupted from the meeting with a virulence that stemmed from the knowledge that he had been outmanoeuvred. (p. 172) The stack shows that the Master’s strategic thinking was evaluated from different perspectives: the Master’s own reflection, Skullion’s observation and the Dean’s reaction. Unlike the Master, initially the Fellows exhibited very little of strategy beyond belligerence (pp. 13, 67, 102). Another aspect of the Master’s characterisation can emerge if we consider the group of metaphors that represent demonstrative gestures or showing-off rather than the direct application of force or violent intervention. The Master is clearly attracted to military parades or manoeuvres. Therefore, the next 82 stack coalescing around the demonstration of martial spirit can be labelled PARADING. (14) Stack PARADING For forty years Sir Godber had marched beneath the banner of so- cial justice, or at least paraded... [H]e had piloted through Parliament a series of bills... (p. 3) ‘The iron fist in the iron glove,’ he murmured to himself, conscious that for the first time in a long career of political manoeuvring and compromise he had at long last achieved a clear-cut victory over an apparently intransient opposition. (p. 117) The Dean erupted from the meeting with a virulence that stemmed from the knowledge that he had been outmanoeuvred. (p. 172) With the silent thought that it was time such relics of the past got their marching orders, the Master led the way into the Lodge. (p. 174) And now with Sir Godber triumphant, and the Senior Tutor, at least privately, admitted the Master’s victory... (p. 293) The two systems of metaphors reveal the superficiality of the Master’s radi- calism and his Machiavellian approach in running debates. 7.3 Metaphor scenarios and the Dons’ mind style If the Master’s motivation often stemmed from the desire to show off his talents rather than to pursue some real and distinctive goal, the Fellows’ strength resided in their beliefs of being chosen for the mission of preserving the ‘amiable’ atmosphere in academy. Their wrongdoing could be justified by the very purpose of their existence, e.g. protection of traditions. The sys- tem of metaphors labelled GUARDIANS illustrates their aspirations. (15) Stack GUARDIANS [T]he Dean felt secure in the knowledge that Carrington was the high priest of nostalgia... Traditions sullied. The old and proven ways under threat. The curse of modernism. The Dean could hear the cli- chés now, rolling off Carrington’s tongue to stir the millions hungry for the good old days... The Dean helped himself to sherry with the air of a man well content, if not with the world, at least with the corner of it over which he was guardian. (p. 223–224) Together, though never in unison, they [=the Dean and the Senior Tutor —LA] had steered Porterhouse away from the academic tempta- tions to which all other Cambridge Colleges had succumbed and had preserved that integrity of ignorance which gave Porterhouse men the confidence to cope with life’s complexities... (p. 284) 83 Your attitude suggests that you regard the College as part of a pri- vate domain of which you are custodians. (p. 291) They were guardians of political inertia and their role was done. (p. 293) Sickly unisex would replace the healthy cheerful louts who had helped to preserve the inane innocence and the athleticism that were his only safeguards against the terrors of thought. (p. 293) It is possible to see that metaphor scenarios target the motivation of the pro- tagonists, which in turns shapes their behaviour and attitudes. There is also an interesting metaphor scenario that mirrors the Dean’s methods. As the main opponent of the Master, the Dean prefers behind-the-back actions, set- ups and blackmails to open confrontation. In open confrontations the Dean invokes fear as he demonstrates his disparagement and superiority. However, this method is not effective with the Master, and the Dean resorts to his mighty allies and harassment. The group of metaphors which emphasises this habit of the Dean’s can be labelled as REMOTE CONTROL WAR or GADG- ETS OF WAR. The stack includes some metaphors that the Dean borrowed from General Cathcart. These metaphors from the source domain of HUNT- ING have been effortlessly integrated into the context of WAR, as they repre- sent organised terror and badgering the victim. (16) Stack GADGETS OF WAR ‘Pity,’ said Sir Cathcart. ‘Useful bait, boys.’ ‘Bait?’ asked the Dean. ‘Bait the trap.’ ‘Trap?’ ‘Got to have a trap. Weak spot. Bound to have one. What?’ said the General. ‘Bleating of the sheep excites the tiger...’ [...]Slowly and dimly, through the shrapnel of Sir Cathcart’s utter- ances, the Dean perceived the drift of his thought. ‘Ah,’ he said. (pp. 188–190) [T]he Dean felt secure in the knowledge that Carrington was the high priest of nostalgia. Sir Godber’s plan would be the bait. Carring- ton would make mincemeat of the man’s pretensions. (p. 223) Throughout it all the Dean remained patently uninterested, ... his mind entertained by the fuse that had been lit in Cornelius Carrington. (p. 225) Both parties to the conflict exhibit an aggressive mind style for which many metaphors are accountable, e.g. belligerent, attack, crusade, being offensive, infighting and others, though each character has a peculiar behavioural pat- tern. Metaphors create systems which highlight their mode of conduct. 84 7.4 ANIMAL metaphors It has been previously shown that a few source domains contribute to the context of animosity introduced by the WAR metaphors. I have argued that WEATHER or CALAMITY metaphors and HUNTING metaphors play a part in the re-creation of the hostility which reigned in this academic institution. Proximity to fire or comparisons with the flame were frequently used to highlight the intensity of anger and hate (pp. 13, 67, 117, 118, 157, 158 and others). However, I would also like to discuss the input of ANIMAL meta- phors. As has been previously demonstrated, there are a number of meta- phors contributing to the characterisation of the Dean as a vicious dog. It is interesting to note that both parties to the conflict draw comparison with animals. For instance, the Master and the Bursar are described as dogs. (17) Like two elderly dogs they circled warily in search of the odour of agreement, sniffing each hesitation for the nuance of complicity. (p. 38) Both parties are often equated through similar parallels. For instance, they exchange the ‘swine’ abuses, though not as effrontery. The Dean calls the Master ‘the swine’ (see example (11), strand 1) while Lady Mary holds the same opinion about the Dean (example (9), strand 1). The main characters, regardless of allegiance, can ‘snarl’, ‘bark’, ‘bite’ and ‘snap’. Apart from the low-level animal mapping there are also unam- biguous conceptualisations of the men as animals. (18) Not since the days of the Protectorate had the Council Chamber of Porterhouse known such vehemence, and the Fellows sat staring at the Master as at some strange animal that had assumed the shape of a man. (p. 291) (19) There was something vaguely barbarous about the Hall, as if it were a shrine to appetite and hallowed by the usage of five hundred years... Sir Godber shuddered at the superstitions they had entertained as if he could undo the thread of time that linked him to their animality... Sir Godber halted in the garden, astonished at the idea that he was the product of such a strange species. They were as remote to him as pre- historic animals and yet he inhabited buildings which they had built... Alarmed at his new apprehension of his pedigree, Sir Godber peered around him in the darkness and hurried down the path to the Master’s Lodge. (p. 227–228) The proliferation of the ANIMAL metaphors may yield a few explanations. 85 1. The ANIMAL metaphors create a symmetry in the characterisation of the Master and his allies and the Dons. It seems to be a sign of the author’s impartiality and his unwillingness to take sides in the con- flict. 2. Both parties, as animals, follow their instincts in finding modes of conduct in their struggle with enemies. The longevity and ferocity of their confrontation are determined by the rules of nature that they unconsciously adhere to. 3. The Master and the Dons are similar in being products of a specific culture of violence. However, they represent different species, e.g. the Master is a genetically modified breed of aggressive and domi- nant creatures. The major problem of the Dons is in their inability to adjust to ‘the climate challenges’ (p. 106) or any changes in their environment. The Dean offsets this deficiency through deviousness and ferocity. The Master can adjust to changes but he does not have an eye for details when it comes to the outer reaches of his battle- fields (e.g. Skullion). The Master and the Dons cannot fill the same niche at the same time. Whatever conceptual perspective has been taken in understanding the conflict, e.g. WAR or STRUGGLE FOR SURVIVAL, the tragic ending is inevitable. 4. The ANIMAL metaphors are coherent with the mappings suggested by the WAR metaphors. They overlap with the WAR concepts and en- rich some aspects of the WAR scenario, e.g. hunting tools, struggle for survival, survival of the fittest and others. In such an environ- ment, to survive is the same as to win. 7.5 Perspectives A number of aspects of cataloguing and grouping instances of metaphors have been left out of consideration. For instance, the issue of boundaries between clusters of metaphors in the novel is worth discussing. The Camer- on & Stelma (2004) approach to the representation of clusters in graphic forms could shed light on the internal composition of metaphor groups, their length and average distance between units. The constitution of clusters, e.g. a detailed account of metaphors from the WAR, ANIMAL, FIRE, IRON and WEATHER source domains may reveal the foregrounded conceptual links between categories. Types of narrative disruption (Attardo 2002) within metaphor clusters could be of interest, too. The analysis of disruptions and group coherence may suggest a structure of interpretative segments, e.g. boundaries of text units that are processed simultaneously. It is quite prob- lematic to decide whether we prioritise information on chief characters or events with timeline boundaries. The cataloguing of instances of humour in Attardo 2001, 2002 resulted in extending our knowledge about variation in humorous acts. Similarly, a 86 more detailed picture on metaphoricity could be drawn from clustering Tom Sharpe’s data. It is interesting to see how dead metaphors have been revived through inclusion into a metaphor group. The origin of dead metaphors is worth considering. For instance, there are a number of metaphors derived from Latin, e.g. to prevaricate, pusillanimous, champion, to succumb. Alt- hough ‘to prevaricate’ derives from the Latin verb meaning ‘to walk crook- edly’, its adaptation within the rich context of war may yield the reading ‘to avoid a frontal attack’, alongside ‘evasive tactics’. If ‘pusillanimous’ literally means ‘having a small mind’ and may allude to the ANIMAL domain via its constituents, the initial use of the metaphor probably has focused on the spir- it of cowardly warriors in battlefields. In the verbal context of Porterhouse Blue the basic scene of the metaphor application has been revived. The word champion also revitalizes its military etymology. Koller 2004: 68 argues: [C]hampion also ties in with war terminology, as it is related etymologically to campaign, both of which derive from the Latin word for ‘field’, campus. The Dons’ familiarity with Latin influences their use of ‘champion’ since the Senior Tutor refers to fighters and protectors of underdogs in (20). This us- age attenuates the common linkage of ‘champion’ with sports and friendly rivalry. (20) We should appear as the champions of the underdogs and the Master would find himself in an extremely difficult position. (p. 286) The word ‘to succumb’ originally referred to the change of position along the vertical axis and is linked to cumbo, ‘to lie’. It may have a visualisation, e.g. the image of being knocked down, and therefore, shows a slight differ- ence from ‘surrender’ or ‘capitulate’. The analysis of etymological ties of metaphors and activation of those ties in contexts uncovers semantic twists, context effects and dynamics of relations between source and target do- mains. The stock of metaphors collected from Porterhouse Blue can be used for estimation of semantic distances between topics and vehicles (Cameron 1999a). I would like to consider a few directions which I am pursuing in my other works. Firstly, the metaphor scenarios retrieved from Porterhouse Blue, e.g. GUARDIANS, and GADGETS OF WAR, appear in other English novels. These scenarios in modern English literature have numerous renditions. This direc- tion of research will assist in uncovering variations within the conceptual systems and general themes which attract these metaphor devices. Second, the reader’s individual response to the metaphors may vary too. Therefore, the same metaphors give rise to different interpretations (Semino & Culpeper 2002a). In other words, the psychological study of reading (Steen 2002: 187) needs to be considered. The metaphors collected from the 87 novel could be used in experiments with readers. The informants can be asked about evaluations and associations evoked by each set of metaphors or individual metaphors. There are specific measures such as lexical decisions, self-report measures and attribution tasks (Hasson & Giora 2007, Gibbs & Boers 2005) which elicit the reader’s interpretations and responses. The se- lection and classification of metaphors from the novel are only a pre- requisite for these further studies. 8 Concluding remarks The point has been made that different readers may prime non-identical sets of metaphors while reading novels (Popova 2002) and therefore, arrive at different interpretations. As Porterhouse Blue reflects upon a war, though a metaphorical one, an immediate and subconscious response to it can be a reader’s attempt to take sides in the conflict. Both parties attract unsympa- thetic remarks from the author and it is hard to decide which party Tom Sharpe would choose to support. The division of opinions in the College seems to be tactical rather than a matter of principles. Yet, the war is a colli- sion between minor reforms and unshakeable traditions. As a reader, I have developed my likes and dislikes among the characters and mind styles they epitomise. Basically, any reader can possess a sympathy either toward stra- tegic thinking and passion for triumphant parades or toward the blind protec- tion of traditions through group coercion. My personal preference goes to strategic thinking, albeit it is not a deeply-rooted one. Despite my apprecia- tion of the Master’s handling of the crises, I cannot ignore the message that instincts and inertia gained the victory at the end and therefore, are generally more beneficial for survival. Though the Master is more flexible and more human than the Dean and the Senior Tutor, his mind style exhibits hidden conflict; cf. example (21): (21) Power did change one, even the power to dominate a group of elderly Fellows in a fourth-rate college. And it was a little victory after all. Sir Godber’s humanity prevailed. (p. 295) The Master’s humanity has its limitations. Sir Godber is deaf to Skullion’s pleas and the lack of flexibility demonstrated in this instance was his fatal error. People were pawns on a chess board in his battles. Perhaps, Sir God- ber’s long affiliation with the power group in politics annihilated his sensi- tivity. War was inevitable for many reasons: a) both parties conceptualise their views in terms of war; b) the Master was a catalyst who stirred up concealed problems (see example (3)); c) the Master would not be able to enjoy peace 88 even if he refrains from interference in college matters. The latter point is illustrated in the next novel, Grantchester Grind: A Porterhouse Chronicle, in which a handicapped and broken Skullion as the Master rebels against the Dean in order to be treated with respect and survive in the Machiavellian games of the Dons. But this is not the only lesson that can be learnt from the analogy of academic life with war. The price of minor changes in a con- servative environment is very high and may take human lives in a literal or figurative sense. Such are my inferences and interpretations of the metaphor systems in my capacity as reader. References Alm-Arvius, Christina. 2008. ‘Iconicity and Poeticity in the Discourse Functions of Figures of Speech’. Paper presented at The 2008 Metaphor Festival, Stockholm, 18 September 2008 as ‘The Discourse Function of Schemes and Tropes’. This volume: 91–125. Attardo, Salvatore. 2002. ‘Cognitive Stylistics of Humorous Texts’, in Semino, Elena & Jonathan Culpeper, (eds), Cognitive Stylistics: Language and cognition in text analysis. Amsterdam: John Benjamins, 231–251. —. 2001. ‘Stylistic Markers of “Serious Relief” in Wilde’s Lord Arthur Savile’s Crime’, Stylistika X: 19–31. Black, Max. 1993. ‘More about Metaphor’, in Ortony, Andrew (ed.), Metaphor and Thought. Cambridge: Cambridge University Press, 19–42. Cameron, Lynne. 1999a. ‘Identifying and Describing Metaphor in Spoken Discourse Data’, in Cameron & Low (1999b): 105–135. —. 1999b. ‘Operationalising “Metaphor” for Applied Linguistic Research’, in Cam- eron & Low (1999b): 3–29. Cameron, Lynn & Graham Low. 1999a. ‘Metaphor’, in Language Teaching, 32(2), 77–96. Cameron, Lynne & Graham Low (eds). 1999b. Researching and Applying Meta- phor. Cambridge: Cambridge University Press. Cameron, Lynne & Juurd Stelma. 2004. ‘Metaphor Clusters in Discourse’, in Jour- nal of Applied Linguistics, 1(2): 107–136. Culpeper, Jonathan. 2002. ‘A Cognitive Stylistics Approach to Categorisation’, in Semino & Culpeper (2002b): 251–279. —. 2001. Language and Characterisation: People in plays and other texts. Harlow: Pearson Education. Deignan, Alice. 2005. Metaphor and Corpus Linguistics. Amsterdam: John Benjamins. Fowler, Roger. 1996 [1986]. Linguistic Criticism. 2nd ed. Oxford: Oxford University Press. Gibbs, Ray W. & Erika Boers. 2005. ‘Metaphoric Processing of Allegorical Poetry’, in Maalej, Zouhair (ed.) Metaphor, Cognition and Culture. Manouba-Tunis: University of Manouba-Tunis, 11–25. Goatly, Andrew. 1997. The Language of Metaphors. London: Routledge. Hamilton, Craig. 2002. ‘Conceptual Integration in Christine de Pizan’s City of La- dies’, in Semino & Culpeper (2002b): 1–23. Hasson, Uri & Rachel Giora. 2007. ‘Experimental Methods for Studying the Mental Representation of Language’, in Gonzales-Marquez, Monica, Irene Mittelberg, 89 Seana Coulson & Michael J. Spivey (eds), Methods in Cognitive Linguistics. Amsterdam: John Benjamins, 302–323. Koller, Veronika. 2004. Metaphor and Gender in Business Media Discourse: A critical cognitive study. Basingstoke: Palgrave Macmillan. Kövecses, Zoltán. 2002. Metaphor. A Practical Introduction. Oxford: Oxford Uni- versity Press. Lakoff, George & Mark Johnson. 1980. Metaphors We Live By. Chicago: Chicago University Press. Leech, Geoffrey N. 1969. A Linguistic Guide to English Poetry. London: Longman. Littlemore, Jeannette & Graham Low. 2006. Figurative Thinking and Foreign Lan- guage Learning. Basingstoke: Palgrave Macmillan. Musolff, Andreas. 2006. ‘Metaphor Scenarios in Public Discourse’, in Metaphor and Symbol, 21(1): 23–38. Popova, Yanna. 2002. ‘The Figure in the Carpet: Discovery or Re-cognition’, in Semino & Culpeper (2002b): 49–73. Scott, Mike & Christopher Tribble. 2006. Textual Patterns: Key Words and Corpus Analysis in Language Education. Amsterdam: John Benjamins. Semino, Elena. 2002. ‘A Cognitive Stylistic Approach to Mind Style in Narrative Fiction’, in Semino & Culpeper (2002b): 95–123. Semino, Elena & Jonathan Culpeper. 2002a. ‘Foreword’, in Semino & Culpeper (2002b): IX–XVI. Semino, Elena & Jonathan Culpeper (eds). 2002b. Cognitive Stylistics: Language and Cognition in Text Analysis. Amsterdam: John Benjamins. Sharpe, Tom. 2001 [1974]. Porterhouse Blue. London: Arrow Books. —. 1996. Grantchester Grind. A Porterhouse Chronicle. London: Pan Books. Steen, Gerard. 2008. ‘When is Metaphor Deliberate?’ Keynote address at The 2008 Metaphor Festival, Stockholm, 19 September 2008. This volume: 47–63. —. 2007. Finding Metaphor in Grammar and Usage. Amsterdam: John Benjamins. —. 2002. ‘Metaphor in Bob Dylan’s “Hurricane”: Genre, language, and style’, in Semino & Culpeper (2002b): 183–211. 90 Christina Alm-Arvius Iconicity and Poeticity in the Discourse Functions of Figures of Speech Abstract: This qualitative study deals with the nature of poeticity and iconicity and their role in the discourse functions of figures of speech: schemes and tropes. The concept of poeticity is that of Roman Jakobson. The poetic function is a particular kind of meaning which is created from language-internal material. It is found in rhythmic schematic repetition and more deliberate tropes whose poetic qualities seem foregrounded and aesthetically designed. Accordingly, they will have rhetori- cal and mnemonic potential. Moreover, poetic uses will have a monistic character, as their form and meaning will fuse, and this may make it difficult to translate and paraphrase them. Metonymic instantiations and conventional, entrenched metaphors will not be noticeably poetic, but the semantic status of a given use will be a result of more specific discourse factors. The poetic function can interact with factually descriptive, affective, and interpersonal meanings, which are extra-linguistically oriented, as well as with meaningful textual structuring. Poeticity is found in many different text types. It will be a global organisational feature in poetry, but tends only to occur locally in prose. In addition, prototypical iconicity concerns motivated similarity between a linguistic form and the kind of phenomenon out in the world that it represents. However, iconicity has also been used about the similarity relation between e.g. a metaphorical meaning and its source. Iconicity and poeticity often occur together, and they will strengthen and help to foreground each other’s charac- ters. Keywords: discourse, figures of speech, poeticity, iconicity 1 Introduction 1.1 Discourse Discourse is language use for some kind of communicative purpose. Any instance of discourse occurs within an extra-linguistic setting, and, in terms of the meaning aspects they can be associated with, discourse occurrences will be complex. They differ as regards temporal and location specificities, and involve different participants, whose relationships and communicative intentions can be of many various kinds. They will be guided by register norms or cultural conventions, some of which differ in crucial respects, and personal or idiosyncratic factors can colour them a great deal. 91 In short, there are many different types of discourse, and particular in- stances of language exchanges can be more or less easy to fit into regularly occurring and generally recognised discourse categories. A brief, oral face- to-face conversation about a practical everyday matter between people who know each other well, like members of the same family, and, for instance, a prolonged attempt by an academic to read and understand a written inscrip- tion from a very distant time and society are different in significant respects. All the same, they are both examples of communication by means of lan- guage, and they will only be successful if a message is transferred. Obviously, functioning discourse depends on the capacity of a language, or its users, to express what we perceive, do, feel, think and imagine. It al- lows communication about human experiences, be they shared or individual, parts of the external environment or inner bodily or mental reactions. 1.2 Figures of Speech Furthermore, language use can be non-figurative or figurative. This distinc- tion is recognised both by language analysts and by language users in gen- eral. It should thus be possible to detect some general distinctive quality for what constitutes figurative language, and which makes it different from non- figurative communication. But importantly enough, this distinction cannot be said to be discrete. Instead there appears to be a continuum between “plain” non-figurative uses and obvious figures of speech, with quite a few indeterminate cases in between. More specifically, the terms figurative and figures of speech have been used both about rhythmic repetitions of linguis- tic forms, schemes, and about non-literal meanings, tropes, and this broad understanding of the notion of figurative language will be applied here. We will focus on prototypical instances of figurative language, but it should be kept in mind that this is not a homogenous and clearly delimited category (cf. Alm-Arvius 2003).1 (1) below exemplifies two types of schemes: constructional parallelism of two verb phrases with complements and alliteration, initial consonant rhyme, on /g/ in give and get and on /l/ in little and lot. (1) Give a little. Get a lot. There is arguably yet another, comparatively inconspicuous scheme in this example: consonance, the rhythmic recurrence of a consonant at the end of or inside words in a language sequence. The stop /t/ occurs both in little and 1 For practical reasons most of the examples in this qualitative study come from written sources, but some of them are based on oral usage, as they are utterances I heard and made a note of, indirect or anecdotal quotations, or tales rooted in an oral tradition. 92 in lot2, and there may even be a potentially noticeable interaction between the /l/ alliteration and the occurrence of this consonant at the end of little.3 (2) His prose is peppered with spicy phrases. In (2) there is also alliteration and word-internal consonance, in this case on /p/. Moreover, there are two intertwined instances of metaphor: “peppered” and “spicy”. Their metaphoricity is clear from the collocation with “prose” and “phrases”. Metaphors signalled by syntagmatic relations are internal, while an external metaphor could in principle be literally understood as well, had not the wider language context or the extra-linguistic situation shown that it was figuratively intended (Alm-Arvius 2003: 78, 115f). The meta- phors in this string result from an imaginative generalisation of the verb pepper and the adjective spicy. When used literally, they represent practices and experiences in the domain of cookery, but here they have been extended to describe a style of writing. The quotation below from Hamlet contains two coordinated metaphorical and also hyperbolic predications: “drown the stage with tears and cleave the general ear with horrid speech”. But the head of the object phrase “ear” is rather synecdochic, as it represents an auditory experience. Such co- occurrence of different analytical categories of tropes is common. (3) Hamlet: … this player here, … He would drown the stage with tears, and cleave the general ear with horrid speech; … (Shakespeare, Ham- let, Prince of Denmark II:ii, 1045) The following Shakespearian chain of oxymora is also hyperbolic. Indeed, such juxtapositions of antonymous lexical items seem also metaphor-like. (4) Duchess of York: Dead life, blind sight, poor mortal living ghost, ... (Shakespeare, King Richard the Third IV:iv, 734) These and other figures of speech involve selective foregrounding of formal or semantic qualities. Figurative uses are compositionally de-signful com- pared to non-figurative language (cf. Leech & Short 1981: 28f). In other words, figures of speech select and highlight particular formal features or semantic potentials of the words used. This can be informatively or rhetori- 2 The /t/ in these two words will however be represented by different allophones in American English. In lot its realisation will be voiceless, but in little a flapped, voiced positional variant will be used. 3 In English English there are two /l/ allophones, light and dark. The first occurs in syllable- initial position, while the velarised, dark variant is found medially or at the end. 93 cally efficient; a snappy, compact, or constructionally cohesive way to pre- sent a message. (5) Drink a pint a milk a day. (advertising slogan and set phrase) The next line, uttered by Mae West in the film I’m No Angel, exemplifies chiasmus, a kind of antithetical parallelism. The order of key words in the first string is inverted in the second one to express an idea in a striking way. (6) Well, it’s not the men in your life that count, it’s the life in your men. In the nursery rhyme stanza in (7) there are several schemes: /w/ alliteration as well as /i;/ and /I/ assonance in “Wee Willie Winkie”, and also end rhyme: “town”—“gown” and “lock”—“o’clock”, and contrastive parallelism: “Up stairs and down stairs”—“Tapping at the window, crying at the lock”. (7) Wee Willie Winkie runs through the town, Up stairs and down stairs in his night-gown, Tapping at the window, crying at the lock, Are the children in their bed, for it’s now ten o’clock? (‘Wee Willie Winkie’ 2006) The sequentially—or “horizontally”—developed schematic rhythm increases textual cohesion and is mnemonic. This explains why schemes are common in traditionally constructed poems, in much oral literature, and in slogans and idioms. By comparison, the figurative meaning of a trope is based on some ordinarily literal source contents. A trope is thus bi-dimensional, or perhaps even multi-layered, with a backgrounded and partly suppressed source that serves only as an underlying resource from which selected features and structural relations are extracted. These figuratively used parts need not be central or even necessary in the understanding of the source. So while schematic foregrounding works through horizontal rhythmic repetition of selected forms, figurative readings come about through an, as it were, vertically layered foregrounding of certain aspects of meaning. (8) For Lincoln, this sense of purpose was indeed the key that unlocked the gates of a mental prison. (Shenk 2005: 126) In (8) the phrase “the key that unlocked the gates of a mental prison” has a metaphorical meaning. In terms of syntax, it functions as the subject predicative complement in a copular clause. The subject, “this sense of purpose”, is abstract, referring to an intuitive feeling, a psychological stance. It coherently connects to “mental” in “a mental prison” as well as to the 94 metaphorical reading of the rest of the complement. As is often the case, the metaphorical character of parts of this sentence is obvious from their co- occurrence with words or expressions with abstract meanings. In (9) “Wall Street” and “America” function as metonymic shortcuts. They exemplify a regular type of polysemy, as names or descriptions of places can refer to the people connected with them, or their activities. (9) Does Wall Street protect America? The apposition of “a rat” and “a cold and scheming serpent” in (10) consti- tutes a mixed metaphor, as no creature, and definitely not a man, can be both a rat and a serpent. Mixed metaphors draw on different source meanings, and experiential domains, but they function well as long as the figuratively used parts are compatible. (10) That man is a rat, a cold and scheming serpent. 1.3 Poeticity Functionally prominent figures of speech will have poetic qualities. The adjective poetic and the noun poeticity characterise language use that in- volves aesthetic, imaginative, and affective qualities. Figurative and figures of speech are related in a similar way. More precisely, characterising lan- guage uses as poetic will entail that they are also somehow figurative. The category of poetic language and that of figurative uses are not identical, however, as the latter category seems to be more inclusive. For instance me- tonymies are ordinarily considered figurative, even if they do not usually appear poetic. Significantly enough, they tend also to be less noticeable than metaphors. (11) The shop refused to refund me. This is no doubt a result of the straightforward experiential relation between a more basic source meaning and a metonymic understanding. The latter will retain a natural, actual, and practical connection with the source. The kind of imaginative tension between source and target that will be part of an obvious or “fully alive” metaphor is not at hand in metonymy. The people running a shop will work within this location. Similarly, football—or soccer—players play on a football pitch, and can be associated with specific parts of it, for instance the midfield, and the kind of free kick called a corner is made from a corner of the pitch close to the other team’s goal. There is also a direct physical motivation for the use of head as a verb in a phrase like “headed the ball”, because the ball is then propelled by a player’s head. Conversion, also known as zero derivation, can 95 be analysed as a kind of metonymic shortcut, and a so-called expansion test can be applied to it to produce a longer, more explicit and literal formula- tion: to head a ball is ‘to use one’s head to direct the ball’ (Alm-Arvius 2003: 99, 155f). Also the particular sense of goal occurring at the end of (12) is metonymic, as it means ‘scoring (in a game) by hitting the ball/puck into the frame (with a net) that is called the goal’. (12) The midfield played well in the first half. Peters kicked a corner to Smith who headed the ball to score a goal. The kind of focusing communicative strategy termed metonymy is related to presupposition and conversational implicature (cf. Saeed 2009: 102–111; Grice 1975).4 These are all terminological attempts to explain that language users mention only selected aspects of what they are talking about. Much is not explicitly expressed in the words used, but is merely taken for granted as shared knowledge of some sort—or possibly indicated by paralinguistic means such as eye contact, facial expressions, gesture or body posture in oral discourse, and pictures or a general spatio-temporal setting in writing. En- coders—speakers or writers—will rely on a host of experiential insights shared with the decoders, their addressees. Much encyclopaedic knowledge that is relevant, or indeed essential, in discourse concerns generally occur- 4 Grice suggests that metaphors can be explained as flouting; that is, they seem superficially not to adhere to the conversational maxims, which are presented as rules-of-thumb for being cooperative and successful in verbal exchanges. In particular, metaphors seem to flout the Maxim of Quality, which says that one should be truthful, because they are by definition false if they are analysed from a strictly literal and truth-oriented point of view. All the same, a metaphor will be intended to be understood, and someone who uses such language does then not breach the Cooperative Principle, a terminological notion for the basic kind of attitude that is required for verbal communication to be a joint, purposeful effort. However, it has also been claimed that even if a conversational implicature is intended, it should in principle be cancellable, or defeasible (Levinson 1983: 114). The reason for this is that such implicatures are not established parts of meanings of words and constructions in the language used. This would appear to pose a problem at least for internal metaphors, because they are signalled by syntagmatic combinations, or collocations, that cannot be used to ex- press a literal meaning. A simple example would be Girls are roses, because no real girl is a member of the literal and concrete category of roses. It is of course possible to reformulate this particular type of metaphor as a simile, Girls are like roses, but the question is whether this means cancelling the implicature, since similes are also figures of speech, not literal, verifiable comparisons. A more elaborate explanation of what a metaphor means seems typi- cally infelicitous. In fact, one definitional feature of metaphors is that they cannot really be rephrased (cf. Davidson 1978: 142–145). By comparison, a metonymic shortcut can be made literal by applying the expansion test, but such a longer and more explicit formulation tends to be unnecessarily detailed, or longwinded. This explains why metonymic uses will be both more economical and communi- catively efficient than comparable literal descriptions; cf. e.g. the standard example of meton- ymy: The kettle is boiling with the literal variant The water in the kettle is boiling. 96 ring experiences that at least older children and adults can be expected to have. In addition, particular discourse events tend to implicitly assume in- sights that are specific and possibly just transient pieces of information from the ongoing communicative event. I have pointed out that language use quite generally has a synecdochic character (Alm-Arvius 2003: 164–167). In a recent paper, Langacker makes a similar observation from the point of view of his Cognitive Grammar framework: …since every experience is unique at the level of fine-grained detail, any commonality that is reinforced and established as a routine is bound to be coarse-grained relative to the specific conceptions giving rise to it. The ab- straction (or schematization) which thus occurs can in principle be carried to any degree (2009: 14). Langacker’s theory can also adequately deal with construals of a synecdoch- ic or metonymic kind. Example (13) spells out the metonymic, or synecdochic, relation between the count noun valentine and St Valentine’s Day, which rather has a proper name status. Talking about sending or receiving a valentine is using clipping in terms of word formation, as the noun valentine is short for valentine card, a compound. Moreover, this example illustrates the tendency in metonymy to background, and with time perhaps even obscure, important qualities in the source. Large parts of the cultural concept of St Valentine’s Daв need not be included in the construal of valentine (card). (13) A valentine is a card you send on St Valentine’s Day to someone you love. All in all, this means that there is a taxonomic—or hyponymic—relation between the categories of poetic language and figurative language such that the former is more specific, while the notion of figurative language is broad- er, comprising also language uses that lack poetic qualities. But the core of the category of figurative language is made up of constructions that are also poetic. The impression that more peripheral members of the figures-of- speech category, for instance most metonymic uses and conventional, “dy- ing” metaphors, tend not to be perceived as poetic allows us to recognise a continuum from i) poetic and prototypical figurative language, over ii) non-poetic and non-prototypical figurative language, to iii) non-poetic and literal language. The two categorial poles of this scale, i) and iii), are clearly distinct, but the middle category, ii), merges into both, but, as it were, in different directions 97 along the continuum. Some inexactness must be recognised in the characteri- sation of all these three categories, but the occurrence of clearly distinct pro- totypical members makes it possible to distinguish them. In other words, this scalar model allows us both to set up different analytical categories and to recognise the gradience or the existence of fuzzy boundaries between some of them. As has been outlined, a scheme is some kind of rhyme or parallelism that functions as an echoic, cohesive pattern within a written text or oral stretch of discourse. (14) Bustopher Jones is not skin and bones— In fact, he’s remarkably fat. He doesn’t haunt pubs—he has eight or nine clubs, For he’s the St. James’s Street Cat! He’s the Cat we all greet as he walks down the street In his coat of fastidious black; No commonplace mousers have such well-cut trousers Or such an impeccable back. In the whole of St. James’s the smartest of names is The name of this Brummell of Cats; And we’re all of us proud to be nodded or bowed to By Bustopher Jones in white spats! (Eliot 1962: 49) The prose extract below is from the beginning of the main text of an article. The preceding introduction reads: “Children’s wild imaginations are the wellspring of an amazing logical skill: the ability to understand how one thing leads to another”. This passage exhibits rhetorical qualities of the kind found in parallelism, but the structural repetitions here are mostly just par- tial, not prototypical or “ideal” examples of this formal, figurative device. (15) Human beings don’t live in the real world. The real world is what actually happened in the past, is happening now, and will happen in the future. But we don’t just live in this single world. Instead, we live in a universe of many possible worlds, all the ways the world could be in the future and also all the ways the world could have been in the past or might have been in the present. These possible worlds are what we call dreams and plans, fictions and hypotheses. Philosophers, more dryly, call them “counterfactuals”. (Gopnik 2010: 36) A trope, on the other hand, constitutes a semantic extension of a more basic meaning—the source—which foregrounds another but still related meaning: the target. Accordingly, a trope is part of a polysemous network, which at least consists of the literal sense and the secondary figurative understanding. This interpretative complexity can also be described from a more directly 98 psychological point of view, as tropes can be said to tie together experiential or cognitive phenomena that are clearly different, as well. It is the partial sharing of semantic, or experiental, qualities in separable concepts or cogni- tive domains that makes it possible to construct meaningful tropes. The use of tropes thus allows shifting perspectives on things we communicate about. Their occurrence supports the claim that language meanings and cognition are intimately—not to say inextricably—intertwined. (16) During the deepest freeze of the Cold War, the U.S. and the Soviet Union embarked on an arms race that raised the spectre of a doomsday war. In a trope the focus is on the figurative target, while the source is back- grounded, but still active as long as the trope is “alive”. In unquestionable tropes these two interpretations are clearly distinct, but they must also be related for a figurative shift to make sense. Moreover, the degree of poeticity in a trope seems to depend on the perceived distinction between a source and an extended, figurative meaning. This distinction will be more obvious e.g. in new or unexpected metaphors, or in contexts where the figurative quality of a use is contextually highlighted. (17) She is a gate-crashing tank. (18) The girl bride tried to silence the whining, whispering fears in her mind. In conventional metaphors, like the ‘understand’ senses of grasp and see exemplified below, habitual users of a language will directly access a ready- made sense, so the association to the source will not be very active. (19) Mary couldn’t grasp the meaning of what he was saying. (20) I didn’t really see the point of the argument. In such obscured or “dying” metaphors the interpretative dependence on the source will have faded away, and the meaning will largely have lost its bi- dimensionality. All the same, it may occasionally be “revived” (cf. Steen, this volume). Arguably, this is the case in (21), where the incidental syntag- matic relation of the non-literal senses of fumble, grope, and grasp appears to bring out their metaphorical character. The reason for this is presumably that they all basically describe bodily movements, and when they are strung together like this, the source conceptions of concrete, directly perceptible physical actions are more easily evoked, even if each of these verbs has a lexicalised secondary sense that represents a type of psychological process (cf. Alm-Arvius 2006; Hanks 2006). 99 (21) Simon was fumbling in the dark, desperately groping to grasp the meaning of this message. Similarly, the typically non-poetic character of metonymic shifts is presum- ably a result of the experiential co-occurrence of the phenomena represented by a more basic source sense and a metonymic meaning. (22) to (24) exem- plify regular types of metonymy, or regular, predictable polysemy patterns. (22) The car braked and hooted. (23) He was forced to swallow the whole spoon. (24) The table burst into laughter and applause. By comparison, a metaphorical extension will reach into another experiential domain than the one that is associated with the literal source contents. Some- times there are several metaphors with different sources in a stretch of dis- course conveying an intricate, commonly abstract line of thought. Such met- aphor complexes may involve the repetitive occurrence of certain words or phrases, a feature that also connects them with the category of schemes. (25) Using shorthand, people often said that Lincoln had two distinct moods. … It’s not that his moods turned in a cycle, as day gives way to night, but that he lived in the night and made a strong effort to bring the sun in. (Shenk 2005: 117). The next example, (26), plays around with the polysemous potential of the adjective loose and the metaphorical idiom pull the аool over somebodв’s eyes. This gives the passage punning qualities. They seem especially notice- able in the last sentence, which involves external idiom breaking and meta- phor reversal (Alm-Arvius 2007: 20f), as the literal source of this idiom is also triggered by the context without a change of its canonical form. In addi- tion, the first two strings exhibit a kind of chiasmus. (26) There’s a girl out there with a loose sweater. I wish there was a loose girl there with a tight sweater. I’d certainly like to pull the wool over her eyes. (Groucho Marx in Green 2005: 178; italics of the last two words in the original) The following text extract illustrates how dense with figurative uses— schemes and tropes—a stretch of language use can be without seeming to deviate from appropriate stylistic norms. They help build the text through its topical perspectives and contribute to its informative (or speech act) force. (27) No matter how many years may pass, the shame of slavery will color our country’s heritage. It was an established fact long before our birth 100 as a nation; it caused our greatest war; it has shadowed every struggle, defeat, and victory of our land. Whites still apologize for it, blacks still resent it, and we are all oppressed by its legacy. (Taylor 2005: ix) As has been outlined, figures of speech come about through the selective or contrastive foregrounding of either specific formal characteristics or certain perceived semantic potentials of elements that belong to a language system. These elements can be phonological, grammatical, lexical, and textual, and they are parts of—or acquired together with—a language, for instance stand- ard English. Accordingly, poetic constructs make use of language-internal resources. However, it is also clear that such a language-based perspective cannot escape its dependence on cognition or encyclopaedic experience. 1.4 Iconicity By comparison, iconic items and patterns seem to reflect things outside a language, things that are part of the non-linguistic world, either the perceiva- ble outer world or our inner cognition. More precisely, iconicity is a charac- teristic of both i) mimetic forms and ii) figurative meanings exploiting a more basic literal content. They constitute two general, distinguishable icon- ic categories, but the iconicity of the former, the mimetic forms, seems to be more generally recognised. However, some tropes, notably metaphors and similes, can also be said to have iconic features, as they build on and reflect selected qualities in the source. The phonological forms of the three italicised verbs in (28) seem motivat- ed by the different types of bird sounds that they denote. This mimicry is by no means perfect, but adjusted to the phonology of English and generally shared, conventionalised pronunciations. (28) Robins tweet, but owls hoot and screech. In (29) an artful and playful and also earnestly love-inspired image of the sun, personified, is projected onto Juliet by Romeo. It shows how complex a metaphorical comparison can be. (29) Romeo: But, soft! What light through yonder window breaks? It is the east, and Juliet is the sun. Arise, fair sun, and kill the envious moon, Who is already sick and pale with grief That thou her maid art far more fair than she. (Shakespeare, Romeo and Juliet II:ii, 912) Iconic forms are found at all organisational levels of a language: in the pho- nology, the grammar, and the lexicon, and even at the textual level. As these 101 are interrelated in the construction and use of a language, the realisation of a formal iconic potential will involve more than one analytical language level. So iconic meanings seem somehow to mimic matters that are outside the language system, and this appears more obvious in iconic forms than in tropes. Poeticity, on the other hand, results from selecting language-internal material in a way that is aesthetically pleasing or rhetorically efficient. Nonetheless, iconic and poetic qualities may be felt to interact, or even be intertwined, in language usage. In the lyrics of a now classic popular song, the phonemes /i;/ and /ɪ/ are rhythmically repeated in the refrain: “itsy bitsy teenie weenie … bikini”. It exemplifies assonance, interconnected with end rhyme and consonance. In addition, these vowels seem to be potentially iconic, and will then be associated with smallness (Jakobson & Waugh 1988: 187f). This reflects the topic of this song text about a girl wearing a bikini, a small article of clothing. The reason for this iconic potential may be that these two vowels are produced with a raised tongue through a comparatively small opening in the mouth, and, also, that they resemble high-pitched sounds heard from small creatures, e.g. mice. (30) It was an Itsy Bitsy Teenie Weenie Yellow Polka Dot Bikini That she wore for the first time today An Itsy Bitsy Teenie Weenie Yellow Polka Dot Bikini So in the locker she wanted to stay (Vance 1960) Through language we can communicate about things we see, feel, and imag- ine, and the co-occurrence of iconic and poetic qualities may enhance the expressive and rhetorical capacity of a message. Clearly, iconicity in language use is to do with encyclopaedic experiences of the environment as well as with related cognitive structures and capaci- ties. These observations can also be connected with the conception of possi- ble worlds, because we humans are not restricted to thinking and talking about phenomena we have actually perceived and real situations we have been involved in. We can also imagine and linguistically communicate about situations that have not yet occurred, or about situations that may never take place, and even about situations that are factually impossible, just some kind of fiction, although they are still conceivable. The analysis of meaning factors involving poeticity and iconicity in vari- ous types of discourse must address the occurrence and character of different language functions. The recognition of factually descriptive meaning, which represents verifiable phenomena out in the world, must be complemented with the examination of affective and interpersonal meaning aspects. These are all extra-linguistically oriented, and are to do with perceptions, activities, reactions, and social interactions that are not primarily dependent on lan- 102 guage items and structures, even if we can communicate about them verbally (cf. Halliday 1996; Jakobson 1996; Bühler 1982). In addition to these three language functions, there is the poetic function, or meaning based on qualities, items, and relations from within a language. It arises through the selection and combination of phonological resources, grammatical forms, lexical items, and syntagmatic constructions. In fact, the producers of discourse or text select particular presentational perspectives. We need thus also speak of a textual function, or textual mean- ing. Poetic and textually-induced meaning aspects are dependent on the lan- guage used, but they interact with the other three extra-linguistically oriented meaning types. The following text extract constitutes the beginning of Rob- ert Louis Stevenson’s well-known story ‘The Strange Case of Dr Jekyll and Mr Hyde’. It exemplifies how factually descriptive, affective, and interper- sonal meaning qualities can be integrated with poetic features in a textual sequence. (31) Mr Utterson the lawyer was a man of a rugged countenance, that was never lighted by a smile; cold, scanty and embarrassed in discourse; backward in sentiment; lean, long, dusty, dreary and yet somehow lovable. At friendly meetings, and when the wine was to his taste, something eminently human beaconed from his eye; something indeed which never found its way into his talk, but which spoke not only in these silent symbols of the after-dinner face, but more often and loud- ly in the acts of his life. (Stevenson 2002: 5) 2 More about Iconicity In the preceding section iconicity was outlined as the perceived similarity of  either a linguistic form and a thing out in the world that is represent-  ed by this form or the target meaning and the source of a trope, e.g. a metaphor. An iconically motivated form can somehow be felt to have the same, or ra- ther similar, characteristics as the extra-linguistic thing that it stands for. This sort of iconicity can be called formal iconicity, and it seems to be the prototypical way of understanding iconicity in language. The most promi- nent kind of iconic linguistic form is probably sound iconicity, which is ei- ther onomatopoetic or sound symbolic. Onomatopoeia occurs when there is a similarity between some phoneme or phonotactic sequence and the kind of audible experience that it represents. (32) The door opened with a click. 103 (33) The car started with a vroom. (34) Suddenly we heard a whizz and a big shell fell just beyond us. (35) Behind us we could hear the clank and clatter of an approaching train. By comparison, the term sound symbolism can be used when there appears to be some other kind of similarity between the pronunciation of a language element and the (type of) experiential phenomenon it stands for. Such in- stances can be felt to reflect the size, the form, or the movement of what is described. (36) Timmelim sounds smaller than tummelum. (37) Ombulo sounds like something round, while takete is a better name for the spiky figure.5 (38) The flags were flickering and fluttering in the wind. (39) The boys were practicing yo-yo tricks. In (40), from Winnie-the-Pooh, “Woozle” is contrasted with “Wizzle”, and the first one sounds bigger. Actually, the pawprints attributed to creatures given these names were made by Pooh and Piglet themselves. (40) “The tracks!” said Pooh. “A third animal has joined the other two!” “Pooh!” cried Piglet. “Do you think it is another Woozle?” “No,” said Pooh, “because it makes different marks. It is either Two Woozles and one, as it might be, Wizzle, or Two, as it might be, Wiz- zles and one, if so it is, Woozle. Let us continue to follow them.” (Milne 1965: 34f) A certain emotive response may also appear to be mirrored or evoked by a particular language sound. The introductory consonant cluster /sl-/ common- ly occurs in words with negative meanings in English. The question is whether this is i) a natural connection between a sound type and an affective reaction, or whether it is a result of ii) a convention within the English speech community. Both these factors may well be involved. (41) He was a slick, sly, and sleazy businessman. Extra-linguistically motivated sound symbolism can be termed natural sound symbolism, while a mere conventional connection between a pho- neme or a phonotactic combination and a type of meaning can be labelled system-bound. The terms primary and secondary sound symbolism have also 5 These are standard examples of sound symbolism, introduced by Wolfgang Köhler in 1927 (Al Hashimi 2007: 60f.; cf. Robins 1971: 13). 104 been used (cf. Attardo 1994: 153). These categories are no doubt fuzzy ra- ther than clearly delimited and discrete. Ferdinand de Saussure, “the father of modern structuralist linguistics”, stated that the relation between the phonological form, or ‘sound image’, and the sense of a linguistic sign—ordinarily a word—is typically arbitrary, i.e. not motivated or iconic (1966: 66–69). Exceptions to this prevailing tenden- cy do occur, but the expression sides, or signifiers, of such onomatopoeic or sound symbolic items will also be constructed according to the phonotactic principles that constrain what sound sequences can occur in a language. They are thus not straightforwardly mimetic, and, significantly enough, such vocabulary units must also be learnt, because their phonological form need not be an unquestionable indicator of their meaning. According to English, the sound a pig makes is oink, but in Swedish it is nöff /n{f;/. So the sound quality of these apparently iconic words is to a large extent the result of linguistic convention. A reason for this may be that pig sounds are difficult to mimic in ordinary verbal language, because they are not similar to occurring phonemes. By comparison, the lowing of a cow is onomatopoeically reflected in similar ways in English and Swedish, as moo /mu;/ and mu /mʉ;/, respectively, and the “characteristic crying sound” of a cat can be spelt meow, miaou, miaow or miaul in English (Collins Cobuild 1995: 922) and mjau in Swedish. Moreover, it is clear that while some cases of sound iconicity are regular parts of a language, other examples of iconic sounds seem rather incidental and optional; that is, merely occasional results of pragmatic or idiosyncratic communicative strategies. Mimetic instances of the latter kind need not be restricted to the phonological inventory of a language. But the distinction and relation between systematic pronunciations of words and imitation of non-linguistic sounds is not simple and clear-cut. The following extract from a story for children on the Internet shows instead how attempts at more di- rect sound imitation can make use of and be mixed with recognisably con- ventional sound images in a language, in this case English. (42) The bird just watched curiously as Max started flapping and leaped from the top of the tree. “A A A A A A A A A A A AHHHHHHHHHHH!!!!!!! PLOP.” … Up to the surface he swam as fast as he could. “GAAASSPPP. OOH!” “Hoom Haaaaaar” laughed the whale, … … VRROOOOMMMMMMM–Splat. The plane and Max crashed to the ground. (‘Let’s Get Energized’, s.d.) 105 As in many other cases, quite precise distinctions between systematic parts of a language and merely pragmatic communicative devices are hard to es- tablish. This is a result of the intricate relation between, on the one hand, language knowledge and usage and, on the other hand, general cognitive capacities, encyclopaedic experiences, and cultural practices. There seems commonly to be a cline between systematic language fea- tures and structures and other types of communicative or expressive behav- iour. It is reasonable to hypothesise that, when a child is acquiring a lan- guage, she is not just learning the systematic core of its phonology, gram- mar, lexicon, and discourse conventions. She will also take after particular ways of uttering something, like prosodic contours and various phonetic realisations, and even the kind of “body language” she sees used with certain expressions. Indeed, it is unlikely that she could learn to use a language without integrating it with quite concrete performative qualities. Iconic adjustment to what is described is in fact an important rhetorical and commonly also aesthetic aspect of especially oral language. A signifi- cant proportion of the communicative force of story-telling, reporting, and acting depends on this (cf. Theune et al. 2006). There are many ways to iconically support factual as well as affective features of something said, and only a few such devices can be mentioned here. However, the in principle infinite variation of such potentials is important for each rendering of a story or when a role in a play is interpreted by an actor in performance. It is possible to emphasise the length or size of something by iconic lengthening of, for example, a liquid consonant and/or a vowel, perhaps to- gether with increased loudness or exaggerated pitch: llooong; liiittle. In a similar way the abruptness of a jump can be iconically marked by dwelling on the release phase of the introductory stop in this word. These sound de- vices can also be supported by iconic gestures, like stretching out the arms while saying llooong or indicating the small size of something that is liiittle by holding up the hand and leaving just a little space between the index fin- ger and the thumb. More global voice manipulation also occurs, for instance in story telling or reading. A whispering voice can indicate that something is done secretive- ly, and sounding angry or threatening can express danger, or, from the oppo- site participant perspective, by a timid or fear-stricken articulation. A special example worth mentioning is the legendary Swedish sports commentator Lennart Hyland. His reporting style was much appreciated, and it included a skilful use of iconic voice qualities. He could mimic the pace of what was going on during for instance a football or ice-hockey match by speaking alternatively slowly and quickly, speeding up or slowing down his voice in synchrony with the players’ movements and handling of the ball or puck. A classic example is how he dramatically followed the puck’s sliding across the ice into the empty Canadian goal in the last but one minute of the ice-hockey final between Sweden and Canada in the Olympic Games in Col- 106 orado Springs in 1962. He managed to simultaneously report on how the puck was sliding into the goal in this decisive last phase of the match by clearly and yet excitedly chaining the words together in a staccato-like way: “Den–glider–in–i–mål” (‘It’s–gliding–into–the goal’).6 So there are different kinds of sound iconicity: the systematic categories of phonological motivation termed onomatopoeia and sound symbolism as well as analytically less easily captured occurrences of pragmatic modulation of the voice to mimic something described. Yet another type of iconic representation concerns whole pieces of dis- course. This will involve narrative aspects, reflecting the temporal and spa- tial structure of the situation or situations reported on in a passage, including the positions and activities of the participants. In other words, the way an oral message or a written text, or part of it, is structured can appear to mirror the setting and the incidents it describes. It can reflect the temporal sequenc- ing of situations, how spatially close or distant things are in relation to each other (proximity), and how much there is of something (quantity), for in- stance some material amount or some temporal or spatial extension. This kind of iconicity has been labelled diagrammatic (cf. Ungerer & Schmid 2006: 300–311; Peirce 1991: 181–183, 249–253, 270). (43) In the Verkhoyansk Mountains of northeast Siberia, Eveny nomads are on the move. Teams of reindeer pull caravans of sledges down the steep slide of a frozen mountain river. Bells tinkle on the lead reindeer while dogs on short leashes dive closely alongside through the snow … (Vitebsky 2005: 3) (44) Booth used the crowd’s roar to cloak the sound as he turned the door- knob of the President’s box. He stepped quietly behind Lincoln and aimed his single-shot derringer, point blank, behind the President’s left ear. The peals of laughter were still echoing as Booth fired, send- ing a .44-cal. bullet tearing diagonally through Lincoln’s brain. As the President slumped forward, Booth leapt from the balcony onto the stage, badly injuring his leg in the process. (Fenton 2009: 114) (45) One day when Little Red Riding Hood’s mother was baking some cakes she was told that the Grandmother was ill in bed. So she took down a basket and filled it with dainties. Among other things she put into it a home-made cake and a little pot of butter. Then she called Little Red Riding Hood to her and said: ‘Little Red Riding Hood, take this basket and go to your Grandmother and bring me word how she is.’ So Little Red Riding Hood set off across the wood, … (‘Red Riding Hood’ 1976: 78f.) 6 Cf. the recording on YouTube (Tokusep 2008). 107 Examples (45), above, and (46), below, are from prototypical versions of well-known fairytales that are part of general folklore, not just in England, but also in culturally connected countries like Germany and Sweden. In the following story there are also examples of sound iconicity and, quite gener- ally, its iconic characteristics are sequentially, or horizontally, arranged in ways that make them function poetically as well. The story is full of repeti- tion in a manner that is common in oral literature. This more or less rhythmi- cal repetitiveness, which is an obvious design pattern in the story, will have served as a mnemonic strategy in addition to being aesthetically and rhetori- cally effective. (46) So the Big Bad Wolf huffed and he puffed and down fell the little house of sticks. The little pigs ran and ran. They ran to the little house of bricks. ‘Let us in! Let us in!’ they called. The little pig in the house of bricks let them in just as the Big Bad Wolf got up to the little house. Rat–tat–tat! ‘Little pigs, little pigs, let me in. Open the door. Do please let me in!’ ‘No, no. We will not let you in. We will not open the door. We will not let you in,’ said the three little pigs together. ‘Then I will huff and I will puff and I will blow your house down!’ ‘You can huff and you can puff but we will not let you in.’ So the Big Bad Wolf huffed and he puffed and he puffed and he huffed—but he did not blow the house down. ‘I will get up on the roof. I will get in!’ said the Big Bad Wolf. And he got up on to the roof. The three little pigs got a big pot of boiling water. The Big Bad Wolf was on the roof! ‘I’m coming down the chimney,’ he shouted. Then he fell, Plop! Into the big pot of boiling water! The three little pigs joined hands and sang Who’s afraid of the Big Bad Wolf! That is the end of the Big Bad Wolf! The three happy little pigs still live in their little house of bricks. (‘The Story of the Three Little Pigs’ 1976: 18f.) A language element or piece of discourse can thus be analysed as iconic, if its form seems somehow to mimic a salient quality or set of qualities in the extra-linguistic phenomenon or situation it is used to describe. Note however that “extra-linguistic” need not mean “real” or “verifiable” or “true of the 108 actual world”.7 It only means that for instance an iconically described situa- tion is conceivable and possible to imagine for someone who knows the lan- guage in question, in our case English. Accordingly, the iconicity of lan- guage descriptions is primarily related to conceptualisations, not to things as they actually are or really happened. As Kant maintained, we do not have direct access to “things in themselves”. Even though the text extracts in (43) and (44) above report on real-world events, (45) and (46) are from fable-like fairy tales where some of the characters are humanised, talking animals. They thus have symbolic qualities, although these need not be intellectually obvious to young children, and their moral, or the complex of experiential insights that they seem to offer, should be relevant for comparable events in the life of hearers or readers of these stories. Perceived instances of formal iconicity are however rarely close-to- identical reflections of the phenomena they represent. The similarity will only be partial, and dependent on convention. More specifically, perceived iconicity will proceed from a restricted perspective, and the suggested simi- larity between a linguistic form and an encyclopaedic experience will have a synecdochic character. In other words, an iconic form will foreground only some selected property of what it describes, but this part-whole connection may still make it possible to intuit a similarity between the form used and the extra-linguistic matter that it represents. Actually, the modernist literary techniques labelled stream of conscious- ness and interior monologue are attempts at mirroring in words an inner flow of thoughts and emotions. (47) is from the very beginning of James Joyce’s novel A Portrait of the Artist as a Young Man, and here we learn about some of the protagonist Stephen Daedalus’s early experiences. (47) His father told him that story: his father looked at him through a glass: he had a hairy face. He was baby tuckoo. The moocow came down the road where Betty Byrne lived: she sold lemon platt. O, the wild rose blossoms On the little green place. He sang that song. That was his song. O, the green wothe botheth. When you wet the bed first it is warm then it gets cold. His mother put on the oilsheet. That had the queer smell. His mother had a nicer smell than his father. She played on the piano the sailor’s hornpipe for him to dance. … (Joyce 1960: 7) 7 In fact, it is arguable that the correspondence theory of truth relies on a kind of iconicity, as it says that a proposition is true if it corresponds to the situation it is intended to describe. 109 The notion of iconicity appears in the main to be associated with mimetically motivated forms, in particular instances of sound iconicity. But the term iconicity is also employed in a broader sense about the similarity between distinct meanings, notably a comparison between a backgrounded, typically literal source and an extended metaphor target. This kind of similarity can be termed meaning iconicity, because it depends on an imaginative and only partial likeness. A metaphor builds on the perceived similarity between a source meaning, or domain, and a typically more abstract target. This simi- larity need not be factually true or verifiable. Instead it may result from af- fective reactions that are connotatively associated to the source content ra- ther than to central and prototypical factual features in the denotatum of the source (cf. Redzimska 2008). Metaphorical uses of lexical items tend to be lexicalised and recorded in dictionaries. They are, however, often language-specific, so such extensions need not follow specifiable universal principles. It is not possible to predict what particular metaphors can occur. All that can be said is that metaphors are motivated by perceived similarity between a source and a target, but the metaphorical potential of words or expressions cannot be determined. The uses of “budding” in (48) and (49) illustrate how a literal source sense and a metaphorical target meaning can be related. The form “budding” occurs in both these examples, but while the first one is an instantiation of the literal sense of this lexeme, the second use of “budding” is metaphorical. (48) You don’t want to tap a budding tree, as the syrup will have poor fla- vour. (Ware n.d.) (49) He was a budding psychopath. Moreover, it is important to realise that metaphorical comparisons need not be limited to the use of individual words. A metaphorical line of thought can extend through a whole sentence or even a longer textual passage. (50) His well of stories never ran dry because he was always refilling it. (Shenk 2005: 115). (51) If I have seen further, it is by standing upon the shoulders of giants. (Sir Isaac Newton) (52) The election of Lincoln ignited the fuse which set off the explosion. The initial detonation occurred in South Carolina, which, within weeks of the election results, declared itself out of the Union, … (Smelser 1971: 113) (53) We may be tossed upon an ocean where we can see no land–nor, per- haps, the sun or the stars. But there is a chart and a compass for us to study, to consult, and to obey. The chart is the Constitution. (Daniel Webster, U.S. Senator, in Theobald & Malen 2000: 95) 110 Examples like these indicate that metaphors do not just tap lexicalised senses within a language, but reach beyond them into general cognition. The obser- vation that a metaphorical comparison can be applied throughout a textual passage suggests that it is not just tied to individual linguistic signs with systematic expression sides and senses. This conclusion is supported by it being generally possible to elaborate and highlight additional aspects of a metaphorical concept in a creative but still structured way. It is thus not sur- prising that it is difficult to paraphrase metaphors in a satisfactory way. In the case of longer metaphorical passages like those exemplified above this seems indeed hardly possible. However, this need not mean that metaphors are merely conceptual, and that language only provides physical and perceptible containment for them. Rather, the fact that metaphors are very common shows that established and systematically structured language senses link on to and dynamically merge with cognition and experience in general. Metaphorical extensions allow language users to draw on habitual linguistic repertoires as well as rich con- ceptual resources, and they can develop both language constructions and cognitive complexes in various ways. Metaphor is an empowering intellec- tual tool, often with affective props, which can help us construct especially abstract reasoning in a dynamic way. Metaphors show how abstractions are rooted in concrete, bodily experiences that we can cognitively reach and handle by means of language. Metaphor spans and combines language and cognition, and also typically relates the concrete with the abstract, using basically concrete images to construct abstract levels of thought. 3 More about Poeticity The concept of poeticity in this study is that of Roman Jakobson: Poeticity is present when the word is felt as a word and not a mere representa- tion of the object being named or an outburst of emotion, when words and their composition, their meaning, their external and inner form acquire a weight and value of their own instead of referring indifferently to reality. (Jakobson 1981: 750) According to Jakobson, the equivalence principle is the basis for the poetic function of language: What is the empirical linguistic criterion of the poetic function? In particular, what is the indispensable feature inherent in any piece of poetry? To answer this question we must recall the two basic modes of arrangement used in ver- bal behaviour, selection and combination. If ‘child’ is the topic of the mes- sage, the speaker selects one among the extant, more or less similar nouns like child, kid, youngster, tot, all of them equivalent in a certain respect, and then, 111 to comment on this topic, he may select one of the semantically cognate verbs – sleeps, dozes, nods, naps. Both chosen words combine in the speech chain. The selection is produced on the basis of equivalence, similarity and dissimi- larity, synonymy and antonymy, while the combination, the build-up of the sequence, is based on contiguity. The poetic function projects the principle of equivalence from the axis of selection into the axis of combination. (Jakobson 1996: 17) Accordingly, the poetic function is a type of meaning that is dependent on language itself, or on the selective use of elements from its formal and se- mantic inventory. It is expressed through rhythmic, schematic repetition, other foregrounding formal devices, and tropes that combine and contrast two or more meanings. The fact that any multi-word utterance or piece of discourse is necessarily sequentially structured is important for linguistic poeticity. It operates within language strings, and also, which is most im- portant, in the cohesive linking of phrasal, clausal, and sentential construc- tions into texts. Not all text-creation is poetic, but the rhetorical and mne- monic force of a text or parts of it will be enhanced by poetic arrangement. Furthermore, it should be kept in mind that poeticity is not just found in poems, but in many different text types or discursive practices, even if it is especially prominent and pervasive in verse. (54) Shall I compare thee to a summer’s day? Thou art more lovely and more temperate. Rough winds do shake the darling buds of May, And summer’s lease hath all too short a date: Sometime too hot the eye of heaven shines, And often is his gold complexion dimm’d; And every fair from fair some time declines, By chance, or nature’s changing course untrimm’d; But thy eternal summer shall not fade, Nor lose possession of that fair thou ow’st, Nor shall death brag thou wand’rest in his shade, When in eternal lines to time thou grow’st; So long as men can breathe, or eyes can see, So long lives this, and this gives life to thee. (Shakespeare, Sonnet 18) Poeticity appears to be a necessary and global feature in any kind of poetry. It is also present in what is called free verse (“free” because it need not use metric patterns and does not rhyme). However, even poems of this kind in- clude formal foregrounding devices that have semantic implications. It may be parallelism or, what is quite common, defamiliarising line breaks or stan- za groupings at unexpected or “unnatural” places from the point of view of 112 standard syntax and prosody (Fowler 1996: 12, 51, 93ff., 115f., 137f.)8. In addition, modern free verse commonly exhibits deliberate or designful uses of tropes. They can also contribute to the defamiliarising, or de- automatising, effect of poetry (Wales 1990: 105f). These features will enrich the associative potential of a poem and lend it its distinctive formal charac- ter, but they tend also to make it more demanding to read (cf. Miall & Kui- ken 1994). “Examination at the Womb-door” is a poem by Ted Hughes in the collec- tion Crow. From the Life and Songs of the Crow. It is organised and devel- oped by the use of parallelism, building up an internal rhythmic structure which is effectfully broken at the end to deliver a poetic punchline and con- clusion. (55) Who owns these scrawny little feet? Death Who owns this bristly scorched-looking face? Death Who owns these still-working lungs? Death Who owns this utility coat of muscles? Death Who owns these unspeakable guts? Death Who owns these questionable brains? Death All this messy blood? Death These minimum-efficiency eyes? Death This wicked little tongue? Death This occasional wakefulness? Death Given, stolen, or held pending trial? Held Who owns the whole rainy, stony earth? Death Who owns all of space? Death Who is stronger than hope? Death Who is stronger than the will? Death Stronger than love? Death Stronger than life? Death But who is stronger than death? Me, evidently. Pass, Crow. (Hughes 1974: 15) 8 Given a general sense, enjambment—‘striding over’—is a near-synonym of defamiliarisa- tion. It denotes the highlighting effect and creative tension that will result from unusual grammatical breaks and pauses in poetry, typically because of idiosyncratic line division (cf. Wales 1990: 146f). 113 In the next poem, ‘September Song’ by Geoffrey Hill, parallelism interacts with idiosyncratic, foregrounding poetic line divisions. It presents glimpses of the remembrance of a young victim in the Nazi holocaust and the poetic persona’s acknowledgement of this horror, and identification with it, as well as the sense of life still going on. (56) born 9.6.32 – deported 24.9.42 Undesirable you may have been, untouchable you were not. Not forgotten or passed over at the proper time. As estimated, you died. Things marched, sufficient, to the end. Just so much Zyklon and leather, patented terror, so many routine cries. (I have made an elegy for myself it is true) September fattens on vines. Roses flake from the wall. The smoke of harmless fires drifts to my eyes. This is plenty. This is more than enough. (Hill 1974: 2442) By comparison, poetic constructions are not strictly necessary in prose, where foregrounding patterning of this kind tends to occur only locally. (57) and (58) exemplify poetic devices in academic texts. Parallelism is a notice- able feature in this introductory passage from a book on the philosophical notion of intentionality: (57) As a preliminary formulation we might say: Intentionality is that property of many mental states and events by which they are directed at or about or of objects and states of affairs in the world. If, for ex- ample, I have a belief, it must be a belief that such and such is the case; if I have a fear, it must be a fear of something or that something will occur; if I have a desire, it must be a desire to do something or that something should happen or be the case; if I have an intention, it must be an intention to do something. (Searle 1983: 1) 114 The italicised part of the short extract below is a slogan-like, pedagogical summary of a scientific claim. It contains end rhyme and parallelism in close combination. (58) Hebb’s concept—actually proposed by Freud sixty years before—was neatly summarized by neuroscientist Carla Shatz: Neurons that fire together wire together. (Doidge 2007: 63) There is something aesthetic about poetic qualities in this broad, Jakobsoni- an sense, and this is connected with their rhetorical and mnemonic poten- tials, and their cohesive effect. Since poeticity is a kind of meaning, it can be analytically contrasted with other types of meaning: factually descriptive, interpersonal, and affective meaning. These latter three meaning types are all extra-linguistically oriented, while poeticity arises from conspicuous figura- tive foregrounding. As has been pointed out and exemplified, poetic fore- grounding occurs i) in rhythmic schematic repetition of phonological ele- ments or syntagmatic constructions, ii) in defamiliarising formal arrange- ments of language strings (especially in poetry), and iii) through the imagi- native selection and promotion of certain source meaning features in tropes. The prose extracts in (59), (60), and (61) below also exemplify schematic repetition and the use of tropes. They again illustrate that when actually cre- ating texts, poetic constructions will be integrated with other, extra- linguistically oriented types of meaning. In particular, poeticity tends to be coupled with or generate affectivity, and this will also have a rhetorical, in- terpersonal effect. (59) The writer of this review … starts by asking an extremely silly ques- tion, and that is, whether or not I have written this book for the pur- pose of giving pleasure to the British child … he proceeds, apparently quite seriously, to make the extremely limited vocabulary at the dis- posal of the British child the standard by which the prose of an artist is to be judged! Now in building this House of Pomegranates I had about as much intention of pleasing the British child as I had of pleas- ing the British public. Mamilius is as entirely delightful as Caliban is entirely detestable, but neither the standard of Mamilius nor the stand- ard of Caliban is my standard. (Wilde 1989: 188) (60) To witness the aftermath of the earthquake in Haiti is to be lost inside a waking nightmare. The markers of this mapless journey are the swarms of looters, children with chopped-off limbs, cities fabricated of sticks and bedsheets, pulverized cathedrals, dogs circling the dead in the streets. Most Haitians have always lived in a society constructed along a narrow ledge on a precipice above the abyss. The rich existed on the plateau above them, unseen in their black-windowed Land Cruisers. 115 Higher still, as if levitating, was the immaculate, blinding white presi- dential palace—the secret desire of all despots—now crushed by the weight of its three Baroque domes. Where the ledge crumbled, the dead cascaded into oblivion. Where it held, people huddled closer, those with next to nothing suddenly with even less. They continue to endure their history—a crescendo of privation and hardships, matched by the strength, pride and dignity. Their nation was born in the con- quest of slavery; it has been shaped by poverty, struggle and faith. (Natchwey 2010: 20) Language messages are for the most part directed at other people. In other words, they have an interpersonal function. This social type of meaning can merge with affectivity, and poeticity tends also to be connected with inter- personal force and affective aspects. The regular connection between such personally oriented meanings and poeticity explains why the latter is rhetori- cally important. Its aesthetic qualities are also tied up with its interpersonal and affective potentials. Poeticity is the meaningful use of aesthetically foregrounded formal and semantic language elements in oral and written communication. Its occur- rence varies among text types and usage tokens. When it is prominent, a monistic analysis is the only defensible one. Such an analysis says that form and meaning fuse so that it is impossible to separate them. This is seen in poems, where formal poetic patterns and idiosyncracies function as more global foregrounding and organisational devices. The monistic character of poems is directly connected with difficulties in translating them (cf. Leech & Short 1981: 15, 19, 24–29). However, even texts that would be categorised as prose can have poetic characteristics merging meaning and form so that at least stylistically fore- grounded parts of them are monistic, and thus difficult to translate. If form is intrinsically interwoven with the expressed meaning, a rendering in another language would be a transposition rather than a translation. The punning instantiations of fly off the shelves and spread one’s аings in the extract be- low from a newspaper article involve both the metaphorical lexicalised sense of each of these idioms and their compositional source meanings. The reviv- ing of the latter is an example of metaphor reversal. (Alm-Arvius 2007: 14– 22; cf. Alm-Arvius 2003: 141–151) (61) Hello, possums! The new pet that’s simply flying off the shelves On the plus side you’ll never have to take him walkies. On the minus side you may have to duck when you walk into the living room. Sugar gliders, you see, like to spread their wings, if you can call them that. 116 A species of possum native to Australia and New Guinea, they have a stretchy membrane across their backs that lets them launch themselves and glide for up to 200ft. And they are flying off the pet shop shelves. With soft brown eyes, fluffy fur and their amusing acrobatics, they are seen as ideal pets. (Salked, Daily Mail, 12 Sept. 2008: 3) This kind of language play on a canonical idiom and its ordinarily back- grounded source is quite common in headlines, advertisements, and jokes. It attracts attention by giving something standardised and well known an addi- tional, non-idiomatic reading that is also easy to access for a proficient speaker of, in this case, English. The style of the passage is informal and humorously worded and also informative in a way that reveals a skilled pro- fessional writer. It contains other poetic devices in addition to the punning idiom breakings outlined above, notably the parallelism with antithetical anaphora linking the first two clauses after the headline. A poetic feature—/f/ alliteration—has a global organising function in the three-page popular science article which is partly quoted in (62). The three words in the title alliterate, and this initial consonant rhyme is echoed within the introductory outline of the contents following it, in the third and last but one paragraph of the introductory section of the main text, and in each of the one-word headings in the following four sections. Furthermore, the standard American English accent is rhotic, so the recurrence of the consonant /r/ within the words in the title is another feature worth noting, in particular as this approximant is found in “freeze” and “fright” as well. /l/ is also partly similar to /r/, because they are both liquids. However, the end rhyme in “fight”, “flight” and “fright”—consisting of the diphthong /aɪ/ and the end consonant /t/—will be an even more striking phonological linking within this mnemonic pedagogical slogan. (62) Formula For Fear When terror rounds the corner and stares you in the face, a powerful force takes charge of your brain. Ancient instincts compel you to freeze, to flee—or to fight off a hungry mountain lion. … When the danger is far away, or at least not immediately imminent, the instinct is to freeze. When danger is approaching, the impulse is to run away. When escape is impossible, the response is to fight back. And when struggling is futile, the animal will become immobilized in the grip of fright. Although it doesn’t slide quite as smoothly off the tongue, a more accurate description than “fight or flight” would be “fight, freeze, flight, or fright”—or, for short, “the four fs”. 117 On a winter morning a few years back, a young woman named Sue Yellowtail went through them all in about 10 minutes. FREEZE … FLIGHT … FRIGHT … FIGHT … (Wise 2010: 5) There are also other figures of speech in this passage, for instance parallel- ism and metaphor. If analysed in more detail, the poems and prose extracts above could help us see how an internal stylistic norm will be established in a stretch of lan- guage use, sometimes only to be rejected later, for instance for some rhetori- cal reason. Especially from a decoder’s point of view this can be compared to priming, as it affects expectations of what is to come. From the encoder’s point of view it will influence what choices are made from the language system. And in a dialogic exchange the interlocutors will, as it were, practi- cally negotiate, or perhaps just unilaterally signal, what stylistic and register features they find useful as the conversation develops. Obviously, the rela- tions between the participants will then be important, for instance their pow- er relations. The extent to which such selective preferences constitute general conven- tions varies between different types of discourse. In registers where verifia- ble factual information will be in focus, like news reporting or technical instructions, there is typically not much demand for, or even appreciation of, poetic presentations, as they may be felt to interfere with the reliability of what is said. But as soon as personal involvement or an ideological stance is allowed to colour text production, affective and interpersonal aspects will be more important, and this may also invite the use of poetic language (cf. Leech & Short 1981: 42–73). A special rhetorical or interpersonal aim may, however, make a writer or speaker disregard the expectations raised by a certain prescribed text type style or general genre norm. Especially in traditional poetry, the use of met- ric verse forms and schematic patterns has been common, but a modernist attitude is rather assumed to encourage functioning deviations from such compositional standards. In some poem categories—for instance sonnets— the global form is rigidly pre-established, while so-called free verse is more open to variation. But, as has been touched on previously in this section, the 118 descriptive label “free verse” is somewhat misleading, as even such poems will be organised by the use of formal foregrounding devices. Since poeticity arises through aesthetic foregrounding of formal or se- mantic material from a language system—or occasionally two or more lan- guages—it has bearing on the validity of the notion of linguistic relativity (cf. Whorf 1956: 52ff; Sapir 1921: 15, 217f). Schematic repetition can create texts that cannot be translated in some respects. Punning will also have a monistic effect, as it depends on a formal expression spanning or combining different meanings, often in a humorous way. Other tropes may also consti- tute translation difficulties. Gideon Toury claims that metaphor can be a major translation difficulty (1995: 81–84, 260–262, 267–277). The following Shakespearian soliloquy is rich in figurative language. There are many bold semantic contrasts, for instance in the form of word combinations that can be categorised as oxymora, and in instances of anti- thetical parallelism: (63) Juliet: O serpent heart hid with a flowering face! Did ever dragon keep so fair a cave? Beautiful tyrant! Fiend angelical! Dove-feathered raven, wolvish-ravening lamb! Despised substance of divinest snow! Just opposite to what thou justly seem’st – A damned saint, an honorable villain! O nature, what hadst thou to do in hell When thou didst bower the spirit of a fiend In moral paradise of such sweet flesh? Was ever book containing such vile matter So fairly bound? Oh, that deceit should dwell In such a gorgeous palace! (Shakespeare, Romeo and Juliet III:ii, 922) The rhetorical—not to say mystical—affective force of schemes can also be seen in magical chants and incantations. Some of these consist of non-words, presumably to signal that they are only understandable to the initiated. They may be exemplified by the famous Disney chant from the Cinderella film: “Salagadoola mechicka boola bibbidi-bobbidi-boo”—even though it seems to lack the frightening quality of the powerful, irrational, and unpredictable unknown that magical chants should supposedly be associated with even today. Witchcraft chants also tend to have kinaesthetic qualities, suggesting rhythmic movement, as in a dance. (64) All (witches): Double, double toil and trouble Fire burn and cauldron bubble (Shakespeare, Macbeth IV:i, 1016) 119 There are also similar lexicalised locutions expressing what people now gen- erally consider superstitious ideas. Still, they retain a position in cultural conceptions and practices. Such magical poetic set phrases can be considered a sub-category of idioms (Alm-Arvius 2003: 181–184). (65) A bride should wear: something old, something new, something bor- rowed, something blue. Many other established expressions voicing a perceived moral wisdom or some generally shared experiential claim are also formulated with the help of rhythmic schematic repetition and striking tropes. Their mnemonic efficien- cy and aesthetic attraction can presumably explain their grip on language users. (66) Once bitten, twice shy. (67) One man’s meat is another man’s poison. Advertising slogans also make use of the persuasive force that will be evoked by poetic language. (68) Andrex toilet tissue is ‘Soft, Strong and Very Long’. (Morrell n.d.). 4 Conclusion In this paper the discourse functions of figures of speech—schemes and tropes—have been looked at from two complementary but also commonly interlaced analytical perspectives, or categorial notions: poeticity and iconic- ity. The term poeticity is used in the Jakobsonian sense, and it is a kind of language function, or meaning. In contrast to the factually descriptive, affec- tive, and interpersonal functions, which concern the experience of extra- linguistic phenomena, poetic meaning primarily arises through selection and combination of language-internal resources. In other words, poeticity fore- grounds features or elements of a language system, in our case standard Eng- lish. The poeticity of prototypical figures of speech is to do with them being aesthetically pleasing, and they are also often mnemonic, with a related rhe- torical potential. These observations—that they are typically more salient, attractive, persuasive, and easier to remember than non-poetic uses—may help explain why poetic constructs seem to be language dependent. Poetic patterns and creative strategies often become entrenched, i.e. stored in the inventory of a language, because they make us pay attention to them. Furthermore, poetic devices and constructions will be part of whole writ- ten texts or oral exchanges. Like other meaningful messages, or aspects of 120 them, poetic constructs will be integrated in and adjusted to typically quite comprehensive texts or pieces of discourse. Accordingly, meaningful textual aspects appear to constitute yet another kind of meaning, which is also in many respects oriented to resources that are stored for use in, or in relation to, a particular language—or, more empirically put, in the linguistic compe- tence and the performance practices and capacities of members of that speech community. All the same, both the poetic and the textual language functions tap into general encyclopaedic knowledge and cognition. This seems to be a prerequisite for them functioning with other extra- linguistically oriented types of meaning. This fusion of meaning and form makes poetic language monistic, and thus difficult both to paraphrase and to translate adequately. Poeticity, or the poetic function, is a particular kind of meaning, and even though it can be analytically contrasted with extra-linguistically oriented types of meaning, it will merge in intricate ways with factually descriptive meaning and, particu- larly, with affective as well as interpersonal factors. The association of poet- icity with affect and interpersonal force is essential for its existence. The rhythmic character of foregrounded schematic repetition will not only be psychologically attractive and prominent; it will also have kinaesthetic as- pects, as it may inspire musical activities, singing, or even dancing. Tropes (for instance metaphor) are also embodied as well as mental constructs, and since they function in verbal communication, they have important cultural connections, as well. By comparison, iconicity is a mirror-like reflection—often far from per- fect—of extra-linguistic matters in language forms, or, less prototypically, a similarity between two meanings: a more basic and usually concrete one and a secondary, figurative content. Language analysts have in the main dis- cussed if, how, and to what extent phonological iconicity exists; that is, on- omatopoeia and sound symbolism. More recently the mimetic quality of larger chunks of language, including whole texts—e.g. narratives—has also been considered. Indeed it seems that the notion of iconicity is generally more applicable to larger language struc- tures than to small phonological units. In future research the association be- tween, on the one hand, denotation and reference and, on the other hand, iconicity should be further analysed. In addition, mimetic aspects of more specific cognitive categories and domains as well as of abstract image sche- mata seem worth exploring. Finally, it should be observed that poeticity and iconicity often function together in language use. Iconic qualities can contribute to poetic fore- grounding, and rhythmic schematic repetition or even the occurrence of tropes may prompt the recognition of iconicity. (69) It had been a rocking, rolling, reeling ride down the road to fame. 121 (70) The second night, along came the hobyahs, creep, creep, creeping. Through the grey gum-trees came the hobyahs, run, run, running. Skip, skip, skipping on the ends of their toes ran the hobyahs. And the hobyahs cried. “Pull down the hut, eat up the little old man, carry off the little old woman.” (Barnes s.d.) References Al Hashimi, Sama’a. 2007. Paralinguistic Vocal Control of Interactive Media. How Untapped Elements of Voice Might Enhance the Role of Non-Speech Voice In- put in the User’s Eбperience of Multimedia. Doctoral thesis. London: Middlesex University. On-line at: http://www.cea.mdx.ac.uk/local/media/downloads/PhDs/ alHashimi/Paralinguistic%20Vocal%20Control_PhDThesis.pdf (retrieved on 8 September 2010). Alm-Arvius, Christina. 2007. ‘Fixed, Flexible, or Fragmentary? Types of Idiom Variation’, in Nenonen, Marja & Sinikka Niemi (eds), Collocations and Idioms 1. Papers from the First Nordic Conference on Syntactic Freezes, Joensuu, May 19–20, 2006. Joensuu: University of Joensuu, 14–26. —. 2006. ‘Live, Moribund, and Dead Metaphors’, in Wikberg, Kay (ed.), Meta- phors. Special issue of Nordic Journal for English Studies 5(1): 7–14. —. 2003. Figures of Speech. Lund: Studentlitteratur. Attardo, Salvatore. 1994. Linguistic Theories of Humour. Humour research 1. Ber- lin: Mouton de Gruyter. Barnes, June (ed.). s.d. ‘The Hobyahs’, in Barking Buddies. On-line at: http://www.barkingbuddies.com/content/view/17/27/ (retrieved on 10 September 2010). Collins Cobuild English Dictionary. 1995. 2nd ed. Sinclair, John et al. (eds). London: HarperCollins Publishers. (Abbr. Collins Cobuild) Davidson, Donald. 1978. ‘What Metaphors Mean’, in Sacks, Sheldon (ed.), On Met- aphor. Chicago: University of Chicago Press, 29–45. Doidge, Norman. 2007. The Brain That Changes Itself. Stories of Personal Triumph from the Frontiers of Brain Science. London: Penguin Books. Eliot, T. S. 1962 [1939]. ‘Bustopher Jones: The cat about town’, in Old Possum’s Book of Practical Cats. London: Faber & Faber. Bühler, Karl. 1982 [1934]. Sprachtheorie. Die Darstellungsfunktion der Sprache. Stuttgart: Gustav Fischer Verlag. Fenton, Matthew McCann. 2009. ‘Murder at the Theater’, in Knauer, Kelly (ed.), Time: Abraham Lincoln. An illustrated history of his life and times. New York: Time Books, 114–115. Fowler, Roger. 1996. Linguistic Criticism, 2nd ed. Oxford: Oxford University Press. Gopnik, Alison. 2010. ‘The Great Pretenders’, in Powell, Corey S. et al. (eds), Dis- cover Presents The Brain. Spring 2010. New York: Discover Magazine, 36–39. Green, Joey (ed.). 2005. Marx & Lennon. The Parallel Sayings. New York: Hyperi- on. Grice, H. Paul. 1975. ‘Logic and Conversation’, in Cole, Peter & Jerry L. Morgan (eds), Syntax and Semantics, Vol. 3: Speech Acts. New York: Academic Press, 41–58. Halliday, M. A. K. 1996. ‘Linguistic Function and Literary Style: An inquiry into the language of William Golding’s The Inheritors’, in Weber, Jean Jacques (ed.), 122 The Stylistics Reader. From Roman Jakobson to the Present. London: Arnold, 56–86. Hanks, Patrick. 2006. ‘Conventions and Metaphors (Norms and Exploitations): Why should cognitive linguists bother with corpus evidence?’, in Cognitive-linguistic Approaches: What can we gain by computational treatment of data? Theme ses- sion at DGKL-06/GCLA-06 (Meeting of the German Cognitive Linguistics As- sociation), 7 October 2006. Munich, Germany, 9–12. Hill, Geoffrey. 1974 [1968]. ‘September Song’, in Abrams, M.H. et al. (eds), The Norton Anthology of English Literature. 3rd ed. Vol. 2. New York: W.W. Norton & Co. Hughes, Ted. 1974. ‘Examination at the Womb-door’, in Crow. From the Life and Songs of the Crow. London: Faber & Faber. Jakobson, Roman. 1996. ‘Closing Statement: Linguistics and poetics’, in Weber, Jean Jacques (ed.), The Stylistics Reader. From Roman Jakobson to the Present. London: Arnold, 10–35. —. 1981. ‘Retrospect’, in Rudy, Stephen (ed.), Roman Jakobson, Selected Writings, Vol. 3: Poetry of Grammar and Grammar of Poetry. The Hague: Mouton, 765– 789. Jakobson, Roman & Linda R. Waugh. 1988. ‘The Sound Shape of Language’, in Rudy, Stephen (ed.), Roman Jakobson, Selected Writings, Vol. 8: Major Works, 1976– 1980. Berlin: Mouton de Gruyter, 1–233. Joyce, James. 1960 [1916]. A Portrait of the Artist as a Young Man. Harmonds- worth: Penguin Books. Kant, Immanuel. 1995 [1787]. Kritik der reinen Vernuft [Critique of Pure Reason]. 2nd ed. Köln: Könemann. Langacker, Ronald W. 2009. ‘Reflections on the Functional Characterization of Spatial Prepositions’, in Korac, Gordana et al. (eds), Belgrade English & Litera- ture Studies. Vol. 1. Belgrade: University of Belgrade. ‘Let’s Get Energized. The Fable of Max’. s.d. On-line at: http://www.solwest.org/archives/Fable_of_Max.pdf (retrieved on 27 August 2010). Leech, Geoffrey N. & Michael H. Short. 1981. Style in Fiction. A Linguistic Intro- duction to English Fictional Prose. London: Longman. Levinson, Stephen C. 1983. Pragmatics. Cambridge: Cambridge University Press. Livingston, Jerry. s.d. ‘bibbidi-bobbidi-boo’. Song Lyrics: Cinderella. On-line at: http://www.fpx.de/fp/Disney/Lyrics/Cinderella.html (retrieved on 6 January 2011). Miall, David S. & Don Kuiken. 1994. ‘Beyond Text Theory: Understanding literary response’, in Discourse Processes 17(3) 337–352. Milne, A. A. 1965 [1926]. Winnie-the-Pooh. London: Methuen. Morrell, Bob. s.d. Reality Training. Selling, Sales and Sales Pitch. On-line at: http://www.realitytraining.co.uk/selling-sales-and-sales-pitch.html (retrieved on 10 September 2010). Natchwey, James. 2010. ‘Out of the Ruins’, in Time, Vol. 175, No. 5, February 8, 2010. Amsterdam: Time Inc., 18–20. Peirce, Charles Sanders. 1991. Peirce on Signs: Writings on semiotics by Charles Sanders Peirce (ed. James Hoopes). Chapel Hill: The University of North Caro- lina Press. ‘Red Riding Hood’. 1976. In Waters, Frank (ed.), Read Me a Story. A collection of stories and verse for the young. New York: Beaver Books, 78–84. Redzimska, Johanna. 2008. ‘Iconicity in Metaphors’, in Johannesson, Nils-Lennart & David C. Minugh (eds), Selected Papers from the 2006 and 2007 Stockholm Metaphor Festivals. Stockholm: Acta Universitatis Stockholmiensis, 131–138. 123 Robins, R. H. 1971. General Linguistics. An Introductory Survey. 2nd ed. London: Longman. Saeed, John I. 2009. Semantics. 3rd ed. Malden, Mass: Wiley-Blackwell. Salked, Luke. 2008. ‘Hello Possums’, in Daily Mail (12 September 2008). Sapir, Edward. 1921. Language. An introduction to the Study of Speech. New York: Harcourt, Brace & Co. Saussure, Ferdinand de. 1966 [1915]. Course in General Linguistic (ed. Charles Bally, Albert Sechehaye & Albert Riedlinger; tr. Wade Baskin). New York: McGraw-Hill. (First published as Cours de Linguistique Générale) Searle, John R. 1983. Intentionality: An essay in the philosophy of mind. Cambridge: Cambridge University Press. Sir Isaac Newton Quotes. s.d. On-line at: http://www.ask.com/questions-about/Sir- Isaac-Newton-Quotes (retrieved on 30 September 2011). Shakespeare, William. 1951. The Complete Works (ed. Peter Alexander). (The Tu- dor Text. The Players Edition). London: Collins. Shenk, Joshua Wolf. 2005. Lincoln’s Melancholв: Hoа depression challenged a president and fueled his greatness. Boston: Houghton Mifflin. Smelser, Marshall. 1971. American History at a Glance. 2nd ed. New York: Barnes & Noble Books. Steen, Gerard. 2008. ‘When is Metaphor Deliberate?’ Keynote address at The 2008 Metaphor Festival, Stockholm, 19 September 2008. This volume: 47–63. Stevenson, Robert Louis. 2002 [1886]. The Strange Case of Dr Jekyll and Mr Hyde and Other Tales of Terror (ed. Robert Mighall). London: Penguin Books, 5–70. ‘The Story of the Three Little Pigs’. 1976. In Waters, Frank (ed.), Read Me a Story. A collection of stories and verse for the young. New York: Beaver Books, 16– 19. Taylor, Yuval. 2005. ‘Introduction’, in Taylor, Yuval (ed.), Growing Up in Slavery: Stories of young slaves as told by themselves. Chicago: Lawrence Hill Books, ix–xxv. Theobald, Neil D. & Betty Malen (eds). 2000. Balancing Local Control and State Responsibility for K-12 Education. Annual Yearbook of the American Educa- tion Finance Association, 21. Larchmont, NY: Eye On Education. Theune, Mariët, Koen Meijs, Dirk Heylen & Roeland Ordelman. 2006. ‘Generating Expressive Speech for Storytelling Applications’, in IEEE Transactions on Au- dio, Speech, and Language Processing 14(4): 1137–1144. On-line at: http://doc.utwente.nl/57674/ (retrieved on 19 May 2007). Tosukep. 2008, 5 September. Den glider in i mål - VM-guld i Colorado Springs 1962. On-line at: www.youtube.com/watch?v=OEeo4GHIzA0 (retrieved on 10 September 2010). Toury, Gideon. 1995. Descriptive Translation Studies and Beyond. Amsterdam: John Benjamins. Ungerer, Friedrich & Hans-Jörg Schmid. 2006. An Introduction to Cognitive Lin- guistics. 2nd ed. Harlow: Pearson Education. Vance, Paul. 1960. ‘Itsy bitsy teenie weenie yellow polka dot bikini’. On-line at: http://www.events-in-music.com/number-one-songs-itsy-bitsy-teenie-weenie- yellow-polka-dot-bikini.html (retrieved on 30 September 2011). Vitebsky, Piers. 2005. The Reindeer People: Living with Animals and Spirits in Siberia. London: HarperCollins. Wales, Katie. 1990. A Dictionary of Stylistics. London: Longman. Ware, Lauren. s.d. ‘How to Collect Maple Sap and Make Maple Syrup. Hobby Sugaring on Your Small Farm or Homestead’. On-line at: http://smallfarm.about.com/ od/orchardsandberries/a/hobbysugaring.htm (retrieved on 28 August 2010). 124 ‘Wee Willie Winkie’. 2006. On-line at: http://www.nursery-rhymes.org/nursery- rhymes/wee-willie-winkie.html (retrived on 6 September 2011). West, Mae. s.d. Film Reference. On-line at: http://www.filmreference.com/Actors- and-Actresses-Va-Wi/West-Mae.html (retrieved on 30 September 2011). Whorf, Benjamin Lee. 1956 [1927–1941]. Language, Thought, and Reality. Selected Writings of Benjamin Lee Whorf (ed. John B. Carroll). Cambridge, Mass: The MIT Press. Wilde, Oscar. 1989 [1891]. ‘Letter to Pall Mall Gazette’, republished in The Fire- works of Oscar Wilde (ed. Owen Dudley Edwards). London: Barrie & Jenkins. Wise, Jeff. 2010. ‘Formula For Fear’, in Powell, Corey S. et al. (eds), Discover Pre- sents The Brain. Spring 2010. New York: Discover Magazine, 5–7. 125 Bárbara Eizaga Rebollar Letting the Cat out of the Bag: On Idiom Use and Representation Abstract. This paper analyzes the concepts underlying idioms from a pragmatic perspective, following the Relevance Theory proposed by Sperber & Wilson (1995). Hence, the distinction between lexicalized and ad hoc concepts is essential for un- derstanding the nature of the concepts underlying idioms and their idiomatic vari- ants. The paper also introduces the idea that concepts expressed by idioms are lexi- calized to different degrees due to their extensive, recurrent and routine use in dis- course, whereas idiomatic variants must be contextually created from the lexicalized concept of the idiom which the variant makes implicitly manifest and which serves as the starting point for the inferential process. Likewise, the way in which concep- tual information is represented and stored in the mind provides an excellent account of how we retrieve and use idioms and their variants in communication: as factual assumptions in the case of idioms whose content is completely lexicalized or as meta-representations in the cases of variants or idioms with partial or attributive knowledge of the underlying concept. Keywords: idiom processing, lexicalized concepts, ad hoc concepts, relevance theo- ry, interpretive and descriptive content 1 Introduction Until not long ago, idioms were considered to be peripheral expressions of language characterized as a fixed and anomalous group of two or more ele- ments with a conventional meaning (Fernando 1996; Flores D’Arcais 1993; Hernando Cuadrado 1990; Moon 1998; Strässler 1982).1 Fortunately, psy- cholinguistics, intrigued by their frequent appearance in discourse, rescued these expressions from the discrimination they had been subjected to for so long. Since then, experiments with idioms have been carried out that proved these expressions were neither dead metaphors nor completely fixed and invariable. Instead, they have shown different degrees of lexical, syntactic 1 I am very much indebted to José Luis Guijarro and Raymond W. Gibbs for constant feed- back and guidance in developing the ideas in this paper, and for their comments on earlier versions. I am also grateful to the audience at the Metaphor Festival 2008 at Stockholm Uni- versity. 127 and semantic flexibility, depending on the relationship between their constit- uent elements and their overall figurative meaning (Cacciari & Tabossi 1988; Gibbs 1992, 1993, 1994, 1995, 1998; Gibbs & Nayak 1991; Gibbs et al. 1989; Glucksberg 1993; Nunberg et al. 1994). In this sense, the cognitive psychologist Raymond W. Gibbs, Jr. has taken their study to its maximum expression, showing the great complexity of the meanings of idioms and the existence of underlying metaphoric concepts motivating them. This paper will present an analysis of idiom processing within a pragmat- ic theory and will show that idioms are processed like any other word or expression, i.e. they are not peripheral expressions with a fixed meaning, as linguists had long thought. Essentially, I will develop Sperber and Wilson’s theory of lexical and ad hoc concepts (1997) and apply it to idiom use and interpretation in discourse. In particular, I will argue that concepts underly- ing idioms and their idiomatic variants present different degrees of lexicali- zation, depending on the extent to which their initially interpretive content has become lexicalized and thus has been encoded by the expression as its new descriptive content. The paper is organized as follows: Section 2 sets out my analysis of idiom use, which distinguishes descriptive, interpretive and metalinguistic uses of idioms. Section 3 sets out a pragmatic account of the concepts underlying idioms and variants. For purposes of exposition, I outline Sperber and Wil- son’s theory of the relation between mental concepts and words. In section 4 I present some remarks on idiom processing and idiom variants, highlighting the interpretive resemblance that has to be inferred for the former and the metalinguistic resemblance that has to be inferred to interpret the latter. In section 5 I argue that the approach advocated here offers a complete view on idiom processing. 2 Idiom use In this section, I have adopted Relevance Theory’s distinction between de- scriptive and interpretive dimensions of language use to explain idiom use and interpretation. The former occurs when the propositional form of an utterance (i.e. its explicature) describes a truthful state of affairs; the latter is employed when an utterance resembles the thought of the speaker or another utterance. If the resemblance is linguistic or formal, then it is metalinguistic; if the resemblance is semantic or logical, then it is interpretive (Papafragou 2000; Sperber & Wilson, 1995; Wilson 1999)2. According to this classifica- tion, idioms form a continuum of cases, ranging from the most descriptive 2 For further discussion of the descriptive and interpretive uses of language, see Sperber and Wilson 1995: 224–231. For a detailed discussion of the notion of metalinguistic use, see Papafragou 2000 and Wilson 1999. 128 uses, generated by fully lexicalized idioms, to the most interpretive and met- alinguistic ones, created by idiomatic variants, as I will analyze in the fol- lowing pages. 2.1 Descriptive uses of idioms Descriptive uses of idioms occur when there is an identity between the con- cept communicated by the speaker and the one encoded by the idiom. The recurrent use of idioms has produced a situation where the concept speakers initially did not endorse as a truthful description has stabilized in memory and so is registered in the lexicon. For instance, the idiom kick the bucket linguistically encodes the concept KICK-THE-BUCKET and thus constitutes the proper descriptive meaning of the idiom, which is to be distinguished from the set of concepts encoded by the words kick, bucket, etc. Let us consider the two following examples: (1) John kicked the bucket because he was very old. (2) John kicked the bucket because he was annoyed. Both (1) and (2) constitute descriptive uses of language, the difference being that in (1) the idiom, kick the bucket, codifies the concept KICK-THE- BUCKET, which is automatically retrieved from memory. In (2), the words kick, bucket individually encode the concepts KICK and BUCKET, which have to be processed once the idiomatic meaning has been discarded. This descriptive use of idioms is even more manifest in the cases of se- mantically ill-formed idioms, such as by and large, where the idiom consti- tutes the only possible interpretation, because the individual concepts of the idiom lack any logical meaning when interpreted together in an utterance. What the descriptive use of an idiom implies is immediate accessibility to the concept of the idiom, at no extra processing cost. 2.2 Interpretive uses of idioms There are intermediate stages of conventionalization, depending on the de- gree to which idioms have acquired their descriptive content. These interme- diate stages may be used in a standardized way and function as generalized pragmatic routines, i.e. sets of stereotypical and meta-represented assump- tions recurrently used and not fully lexicalized (Papafragou 1995, 1996). This brings about an increase in the accessibility of the idiomatic content, which “does not have to be computed but just retrieved from memory; […] immediate and standard effects are yielded by minimized processing” (Pa- pafragou 1996: 172). An idiom whose meaning is in an intermediate stage of lexicalization will have to be interpreted by the hearer, in order to determine its contextual 129 meaning from the idiom provided by the speaker in the utterance. Thus, there is no complete identification between the concept of the idiom and the one the speaker communicates—the propositional form of the idiom is an interpretation of another thought—, because hearers must meta-represent an attributed utterance or thought to understand the speaker’s meaning. The attributed utterance or thought of the speaker resembles the idiom of the utterance, either formally or semantically (Papafragou 2000; Wilson 1999). As stated above, the former is called metalinguistic resemblance, the latter interpretive. The more implications they have in common, the more they resemble one another. Consider the following example: (3) The bank treasurer knew he had cooked his goose when he stole the bank’s funds.3 In (3) the meaning of the idiom cook one’s goose is enriched in the context. So, we know the reason why the bank treasurer’s good name was destroyed and who the referent of one is, the bank treasurer. The interpretation of (3) is quick and does not imply an extra processing cost, because, for the hearer familiar with this idiom, cook one’s goose makes automatically accessible a set of hypotheses, recurrently used, which constitute the descriptive content of the idiom and from which the speaker’s meaning will be derived. In (3) the concept of the idiom the speaker communicates very closely resembles the conventional idiom appearing in the utterance. Thus, idioms like the one in (3), whose meaning is not fully lexicalized, can be treated as generalized pragmatic routines which make the content of the idiom immediately acces- sible at no extra processing cost. 2.3 Metalinguistic uses of idioms As mentioned above, idiom variants do not only resemble idioms semanti- cally, but also formally (i.e. metalinguistically). Creative uses of idioms of- ten, if not always, imply metalinguistic resemblance. Consider the following example: (4) I can assure you that Bill hasn’t spilled a single bean about Mary’s trip to San Francisco.4 To be able to interpret (4), not spill a single bean, the hearer must recognize the metalinguistic resemblance the speaker has established between the idi- om variant in (4) and the attributed form of the implicit idiom, to spill the 3 Example taken from Makkai (1987). 4 Idiom variant taken from McGlone, Glucksberg & Cacciari (1994). 130 beans, which the speaker assumes to be mutually manifest for both hearer and speaker. Thus, the speaker in (4) is echoing the form of the original idi- om, spill the beans, which she implicitly attributes to the hearer, while sim- ultaneously endorsing (4). The variation of the conventional form of the idiom obliges the hearer to assume that the speaker is implicitly echoing the form of the idiom, spill the beans, but not echoing its content. The metalinguistic use of idiom variants shows a difference in the com- municative intention of the speaker and in the weakly manifest effects achieved from its processing or in the way in which they are achieved. In cases like (4), the speaker conveys a dissociative attitude from the attributed form of the idiom, but not from its content. Hence, the hearer must not only interpret the implicit attribution that the speaker is making but also her dis- sociative attitude to the original idiomatic form. The interpretive and metalinguistic uses of idioms and variants imply a larger processing effort, offset by the emotional effects (emphatic, rhetorical, mocking, etc.) derived from their processing. That is to say, the resulting number of effects must be directly proportional to the processing effort the hearer makes to interpret the concept of the variant or idiom used interpre- tively or metalinguistically. Thus, the more difficult it is to establish the rela- tionship of resemblance between the idiom variant the speaker intends to communicate and the idiom implied, the more processing effort the hearer will have to make to interpret the utterance and the higher the number of weakly communicated effects he will derive. Interpretive and metalinguistic uses of idioms and variants, if extensively and recurrently used for a long period of time, may become descriptive, be- ing stored in long-term memory. The passing of the interpretive or metalin- guistic use of an idiom to a descriptive use implies first a pragmatic change and then a semantic one. This gradual and semantic change is motivated by two factors: accessibility and processing cost. The semanticization brings along with it the immediate accessibility of the lexicalized concept, which does not need to be processed anymore but is automatically retrieved from memory at no extra cost, as abundant psycholinguistic evidence proves (Gibbs 1994, 1995; Gibbs et al. 1989). Thus, idioms used interpretively or metalinguistically may gradually acquire a descriptive content. 131 3 Concepts underlying idioms The distinction between descriptive and interpretive or metalinguistic uses of idioms and variants highlights the existence of two types of underlying con- cepts: lexicalized and stable concepts in the mind, linguistically encoded by idioms, and ad hoc concepts, which are temporary, non-lexicalized represen- tations created contextually in the short-term memory during the utterance interpretation process, from the knowledge already stored in the long-term memory (Barsalou et al. 1993; Sperber & Wilson 1997), as in the case of idiom variants and interpretive uses of idioms. 3.1 Mental concepts and words When analysing the relation between mental concepts and words, Relevance Theory states that most words function as pro-concepts, that is, concepts which have to be contextually specified and which are not lexicalized. As they argue (Sperber & Wilson 1997: 3): What we will argue is that, quite commonly, all words behave as if they en- coded pro-concepts: that is, whether or not a word encodes a full concept, the concept it is used to convey in a given utterance has to be contextually worked out. The same word can express many different concepts without all of them being lexicalized in the mental lexicon. For instance, the word open can be used to convey the meaning of ‘uncork a bottle’ as in (5): (5) Open the bottle. or the meaning of ‘open the lid of a machine’, as in (6): (6) Open the washing machine. Hence, this approach considers that the mappings between concepts and words are neither exhaustive (as not every word corresponds to a concept and vice versa) nor exact, but partial. The concepts used in our thinking are much more flexible, richer and varied than the lexical concepts encoded by words. Thus, most mental concepts do not map onto words. Words only seem to be pointers to the concepts involved in the speaker’s meaning (Sper- ber & Wilson 1997: 3–7). 132 3.2 Lexicalized concepts underlying idioms The reasons that have led me to adopt the view of the existence of lex- icalized concepts underlying idioms (though their degree of lexicalization may vary from idiom to idiom and from person to person) are the following: 1. The psycholinguistic evidence on idioms reflects a great advantage in the retrieval speed of the idioms’ meaning compared with that of the variants (McGlone et al. 1994). This evidence seems to support the assumption that idioms are lexicalized (or, at least, present a high degree of conventionalization) in long-term memory, while variants are contextually created. 2. The fact that idioms are used descriptively indicates that concepts motivating this kind of idioms have been lexicalized due to their ex- tensive, recurrent and routine use in discourse, constituting stable concepts in the minds of the speakers. However, this assumption does not imply that what is stable for the speaker must also be so for the hearer. The specific lexicalized concepts stored in the mind of a given speaker or hearer vary from person to person. 3. The range of standard assumptions communicated is, in the case of idioms, quite stable across contexts, i.e. they remain much the same. Again, this does not imply that speaker and hearer have an identical set of assumptions for the concepts expressed by the idioms. What it means is that speaker and hearer operate on the basis of a very ac- cessible set of mutually manifest assumptions, recurrently used. Such a set provides immediate and standard effects yielded by a minimal processing effort, even though the set of assumptions in the mind of the speaker may not be identical to those of the hearer. Likewise, the fact that their content is stable across contexts does not mean that idioms are insensitive to the contexts in which they are processed. The order of accessibility of the assumptions in the ency- clopedic entry of the idiom depends, on the one hand, on the context in which the idiom is processed and, on the other, on the hearer’s in- terpretation. 4. The fact that there are idiomatic variants whose concepts are contex- tually created seems to point to the existence of some general lexi- calized concepts of idioms, from which ad hoc concepts can be built. The interpretations that idioms make accessible are usually schematic and standard. Thus, most of the time they need to be contextually enriched by the hearer to determine the speaker’s meaning. 133 3.3 Ad hoc concepts underlying idioms The interpretive and metalinguistic uses of idioms and variants imply the creation of an ad hoc concept, i.e. a concept that has to be contextually built. The point of departure for the creation of that new concept is the lexicalized concept of the idiom the variant refers to. Let us consider first the ad hoc concepts in idioms, such as (7): (7) Susan racked her brain trying to guess whom the valentine came from.5 In (7), the lexicalized content of the idiom rack one’s brain—‘to make a great mental effort or to try hard to think of something’—may be too sche- matic to communicate the meaning that Susan is trying to guess who has sent her the valentine card. Thus, she will have to contextually enrich the sche- matic meaning of the lexicalized concept underlying the idiom rack one’s brain to be able to interpret (7); in other words, she will have to create an ad hoc concept to interpret (7). In this case, the idiom rack one’s brain makes immediately accessible a set of mutually manifest assumptions about the mental effort a person must make to remember or to guess something. The hearer will enrich that set of assumptions to specify the contextual meaning of ‘guessing the identity of the sender of an anonymous card to Susan’. The creation of the ad hoc con- cept in (7) does not presuppose a great mental effort, because the shared logical implications between the idiom the speaker communicates and the original idiom impedes an exhaustive exploration of the context. The effi- cient and almost effortless creation of the ad hoc concept in (7) is offset by the reduced set of standard and stereotypical assumptions that forms such an ad hoc concept. However, the creation of an ad hoc concept does not always imply quick accessibility with almost no processing effort. Some interpretive or metalin- guistic uses require greater processing effort, as in the case of (8): (8) Mary has rejected a $500,000 job at Anderson & Co. She’ll never find another one like that. She’s missed the whole fleet.6 In order to derive the ad hoc concept, it is first necessary for the hearer to identify the linguistic expression used by the speaker, i.e. the lexicalized concepts that appear in (8)—FLEET and WHOLE—are not the concepts she wants to communicate, but concepts that resemble the idiom miss the boat, 5 Example taken from Makkai’s Dictionary of American Idioms (1987). 6 Idiom variant taken from McGlone et al. (1994). 134 implicit in (8). After disambiguating and assigning referents, the relevant unarticulated constituents of (9) are recovered: (9) Mary has rejected a $500,000 job at [the company] Anderson & Co. Mary will never [in her life] find another [job] like that [offered by the company Anderson & Co]. She [Mary] has missed the whole fleet [by rejecting a job like the one Mary has been offered]. As a result of disambiguating, the hearer may notice that the concepts com- municated by (9) formally resemble the idiom to miss the boat. They share some formal implications with the latter: this variant exhibits a metalinguis- tic resemblance to the attributed implicit idiom. In (9), what the speaker communicates is not that Mary has lost an opportunity, but a much more specific concept that expresses that the magnitude of the lost opportunity can be compared to the effect that it would cause to miss a whole fleet. To interpret this utterance, the hearer establishes a metalinguistic resem- blance between the ad hoc concept—MISS-THE-WHOLE-FLEET in (9)—and the implicit idiom, missed the boat. So the hearer first loosens the concept MISS-THE-BOAT in (9) to make reference to having missed a very good op- portunity and then forms the concept MISS-THE-WHOLE-FLEET, which makes reference to the meaning of having missed an extraordinary opportunity. Besides the loosening process, it is also necessary to enrich the contextual meaning of the concepts of the utterance. In this case, the enrichment re- quires the rejection of the formal use or some aspect of the idiom to miss the boat. This rejection will bring about the parodic and hyperbolic tone of the idiom in (8). 4 Some concluding remarks on idiom processing and idiom variants In the case of the idiomatic variants, the ad hoc concept is created from the relation of metalinguistic resemblance that the speaker establishes with an idiom, which the variant makes implicitly manifest, and from some contex- tual information. The concepts of the idioms from which variants are created correspond to what is called generalized pragmatic routines, as already ex- plained in connection with the interpretive uses of idioms 7. The more famil- iar and extended a routine is among the speakers of a community, the more possibilities there will be of it being creatively used, because the speakers will be able to determine the most accessible aspects of the concept underly- ing the idiom and their relation to the individual concepts underlying the 7 See section 2.1 of the present paper. 135 constituent parts of that idiom, and thus be able to predict its possible seman- tic extensions. For instance, the familiarity of the speakers with the idiom bark up the wrong tree, which refers to the idea of using the wrong method, leads people to consider the wrong aspect more salient than the other aspects of this idi- om. Thus, speakers exploit this aspect to modify the form and meaning of the idiom, as in bark up the worst tree, bark up the right tree, etc. This familiarity with a specific routine indicates that the relation of re- semblance the speaker establishes depends on the degree of mutual mani- festness of the idiom. Thus, the more mutually manifest the resemblance between the variant and the idiom from which it derives is, the less creative a variant will be. In these cases, the resemblance will be stronger, the set of mutually manifest assumptions will be more accessible and thus, the contex- tual exploration will be less exhaustive. By contrast, the more difficult it is to establish a relation of resemblance, the more exhaustive the contextual ex- ploration will be and the more creative the variant will also be. Such extra processing effort is offset by the increase in the number of effects derived from its processing. For example, it is easier to understand an idiom such as bark up the right tree than one such as be off the chain. The former makes the idiom bark up the wrong tree very accessible, from which an ad hoc concept can be derived quite easily, without an exhaustive exploration of the context. However, the latter does not make mutually manifest the idiom from which it was derived or it may not even be derived from any idiom—but be a new one—, so the hearer has to explore the context more exhaustively to find an interpretation that satisfies his expectations of relevance. A further aspect to highlight in the ad hoc concept formation of the vari- ants is that they resemble an idiom metalinguistically, i.e. the speaker echoes an idiom, implicit in the utterance, with a dissociative attitude. Such an atti- tude is communicated as part of the speaker’s meaning. Thus, idiomatic var- iants convey a certain tone of parody, by which the speaker dissociates her- self not (only) from the content attributed to the original idiom but from its form. Finally, for some idioms, it may not be necessary to build an ad hoc con- cept, because the lexical concept is the one literally encoded by the idiom, as seen in 2.1. For instance: (10) Old Mrs. Jennings kicked the bucket before her ninety-fourth birth- day. In this case, interpreting this utterance amounts to disambiguating the con- cept KICK-THE-BUCKET, distinguishing it from the concepts KICK & BUCKET, and then recovering the unarticulated constituents from the utterance, such as 136 the time span of the utterance. The creation of a new concept to interpret this utterance would be a redundant step in this case; it would only mean an extra processing cost. This is likely to happen with fully semanticized idioms, which are recovered like factual assumptions. If the concept expressed by a speaker does not correspond to the lexical- ized concept—as is the case in (7) or (8)—, then it will be necessary to con- struct an ad hoc concept, starting from the lexicalized concepts of the utter- ance, by disambiguating, assigning references, enriching and loosening con- cepts. The resulting concept may be looser, narrower or a combination of both. 5 Conclusions In summary, my proposal aims at providing a pragmatic alternative for the processing of idioms. I hope that the analysis presented in this article will find applications in future pragmatic work and in other fields such as psy- cholinguistics, sociolinguistics, literary or discourse analysis, as it takes into account pragmatic, psychological and cognitive aspects. Although the present account needs more fleshing out, I hope it will at a minimum contribute to the difficult task of figuring out the processes in- volved in interpretation, especially those underlying idioms and their vari- ants. References Barsalou, Laurence W., Wenchi Yeh, Barbara J. Luka, Karen L. Olseth, Kelly S. Mix & Ling-Ling Wu. 1993. ‘Concepts and Meaning’, in Beals, Katharine, Gina Cooke, David Kathman, Karl E. McCullough, Sotaro Kita, & David Testen (eds), Chicago Linguistics Society 29: Papers from the parasession on concep- tual representations. Chicago: Chicago Linguistic Society, 23–61. Cacciari, Cristina & Patrizia Tabossi. 1988. ‘The Comprehension of Idioms’, in Journal of Memory and Language 27: 668–683. Fernando, Chitra. 1996. Idioms and Idiomaticity. Oxford: Oxford University Press. Flores D’Arcais, Giovanni B. 1993. ‘The Comprehension and Semantic Interpreta- tion of Idioms’, in Cacciari, Cristina & Patrizia Tabossi (eds), Idioms: Pro- cessing, structure and interpretation. Hillsdale, NJ: Lawrence Erlbaum Associ- ates, 79–96. Gibbs, Raymond W. Jr. 1998. ‘The Fight Over Metaphor in Thought and Language’, in Katz, Albert N., Cristina Cacciari, Raymond W. Gibbs Jr. & Mark Turner (eds), Figurative Language and Thought. Oxford: Oxford University Press, 88– 118. —. 1995. ‘Idiomaticity and Human Cognition’, in Everaert, Martin, Erik-Jan van der Linden, André Schenk & Rob Schreuder, Idioms: Structural and psychological perspectives. Hillsdale, NJ: Lawrence Erlbaum Associates, 97–116. 137 —. 1994. The Poetics of Mind. Figurative Thought, Language, and Understanding. Cambridge: Cambridge University Press. —. 1993. ‘Why Idioms Are Not Dead Metaphors’, in Cacciari, Cristina & Patrizia Tabossi (eds), Idioms: Processing, structure and interpretation. Hillsdale, NJ: Erlbaum Associates, 57–78. —. 1992. ‘What Do Idioms Really Mean?’, in Journal of Memory and Language 31: 485–506. Gibbs, Raymond W. Jr. & Nandini P. Nayak. 1991. ‘Why Idioms Mean What They Do’, in Journal of Experimental Psychology: General 120(1): 93–95. Gibbs, Raymond W. Jr., Nandini P. Nayak & Cooper Cutting. 1989. ‘How To Kick The Bucket and Not Decompose: Analyzability and idiom processing’, in Jour- nal of Memory and Language 28: 576–593. Glucksberg, Sam. 1993. ‘Idiom Meanings and Allusional Content’, in Cacciari, Cristina & Patrizia Tabossi (eds), Idioms: Processing, structure and interpreta- tion. Hillsdale, NJ: Lawrence Erlbaum Associates, 3–26. Hernando Cuadrado, Luis A. 1990. ‘Sobre Las Unidades Fraseológicas En Español’ [On Phraseological Units in Spanish], in Álvarez Martínez, Mª Ángeles (ed.), Actas Del Congreso De La Sociedad Española De Lingüística. XX Aniversario (Tenerife, 2–6 De Abril De 1990) [Proceedings of the Twentieth Anniversary Conference of the Spanish Linguistic Society (Tenerifa, 2–6 April 1990)]. Ma- drid: Gredos, 536–547. Makkai, Adam. 1987. A Dictionary of American Idioms. New York: Barron’s. McGlone, Matthew S., Sam Glucksberg, & Cristina Cacciari. 1994. ‘Semantic Productivity and Idiom Comprehension’, in Discourse Processes 17: 167–190. Moon, Rosamund. 1998. Fixed Expressions and Idioms in English: A corpus-based approach. Oxford: Clarendon Press. Nunberg, Geoffrey, Ivan A. Sag & Thomas Wasow. 1994. ‘Idioms’, in Language 70(3): 419–538. Papafragou, Anna. 2000. ‘On Speech-Act Modality’, in Journal of Pragmatics 32: 519–538. —. 1996. ‘On Metonymy’, in Lingua 99: 169–195. —. 1995. ‘Metonymy and Relevance’, in Working Papers in Linguistics 7: 141–175. Sperber, Dan & Deirdre Wilson. 1997. ‘The Mapping between the Mental and the Public Lexicon’, in UCL Working Papers in Linguistics 9: 1–20. —. 1995. Relevance: Communication and Cognition. 2nd ed. Oxford: Blackwell Publishers. Strässler, Jurg. 1982. Idioms in English: A Pragmatic Analysis. Tübingen: Gunter Narr. Wilson, Deirdre. 1999. ‘Metarepresentation in Linguistic Communication’, in Work- ing Papers in Linguistics 11: 127–162. 138 Shelley Ching-yu Hsieh and Elena Kolodkina Emotion Expressed with Eyes and Hands in Three Languages Abstract. This paper focuses on fixed expressions containing terms for the body parts eye(s) and hand(s) in Mandarin Chinese, Russian and English. The Sapir- Whorf Hypothesis (Carroll 1977, Whorf 1956) and the theory of conceptual meta- phor (Lakoff & Johnson 1980) provide our theoretical background. We found that the emotion expressed through these expressions is a function of the body part. Be- sides the basic mapping EYE/HAND STANDS FOR EMOTION, we revealed the metaphor- ical mappings EYE/HAND HAVING A PHYSIOLOGICAL EFFECT STANDS FOR EMOTION, EYE/HAND MOVEMENTS STAND FOR EMOTION, and the metonymical mapping EYE/HAND STANDS FOR PERSON EXPERIENCING EMOTION. The concepts ‘eyes’ and ‘hands’ are integral parts of the studied languages, revealing a shared dimension of the mind, and presenting thought in action. The human body provides the same linguistic material to speakers of these different languages, who then project their own cultures and life experiences on the words representing them, thus creating interesting language diversities. Keywords: body-part expressions, Chinese, Russian, English, Cognitive semantics, body as emotion 1 Introduction The eyes and the hands are not only important external body parts, but ex- pressions for them are also used in languages as vehicles for extended mean- ings. This study focuses on compounds and idiomatic expressions containing body-part terms for eye(s) and hand(s) in Mandarin Chinese, Russian and English. Examples taken from four corpora form the database of the present study, thus enabling a cross-linguistic comparison. The four corpora are: Academia Sinica Balanced Corpus of Mandarin Chinese, The National Cor- pus of the Russian Language, The British National Corpus and The Ameri- can National Corpus. We rely on the theoretical perspectives of the Sapir-Whorf Hypothesis (Carroll 1977, Whorf 1956) and conceptual metaphor (Lakoff & Johnson 1980). Corpus evidence provides a cross-linguistic perspective and allows us to test the hypothesis that our bodily experience is connected with the con- ceptualization of different emotions, mental faculties and cultural values. 139 The structure of the paper is as follows: (1) Introduction, (2) literature re- view and theoretical framework, (3) the presentation of the Chinese, Russian and English data under the categories of emotion, (4) a discussion of the emotion concepts shown thereby, (5) conclusion. 2 Literature review and theoretical framework There is a good deal of research on language concerning body parts (Hsieh 2009, 2008, Lin 2003, Tsao, Tsai & Liu 2001, Heine 1997, Hollenbach 1995, Tsai 1994, Heine, Claudi & Hünnemeyer 1991, Sweetser 1990). This section presents studies that focus on cross-linguistic comparison and studies that concentrate on the body parts which our research project is oriented towards. The theoretical assumptions of the Sapir-Whorf Hypothesis (Carroll 1977, Whorf 1956) and conceptual metaphor (Lakoff & Johnson 1980) will then be introduced. 2.1 Research on body-part expressions Specialists in the field of literary studies have investigated how body-part expressions are used to create portraits or characterizations of people in fic- tion, and how such expressions in Russian have changed from Old Russian manuscripts to modern literature (Yerofeeva & Shamsutdinova 2006). According to the Swiss linguist Bally (1951), the imperfection of the hu- man intellect is reflected in our aspiration to spiritualize the environment. It is difficult to imagine what it is like to be dead and lacking a heart, so that the human imagination constantly gives life to inanimate objects. Moreover, a human being will constantly attribute to all the objects in the reality sur- rounding him/her features and aspirations that are characteristic of his per- sonality. Aitchison (1994: 153) states that “body metaphors are … partly based on genuine physical features, partly on convention, partly on imagination”, and that “universal and cultural aspects of metaphors are … intertwined …. Any language selects one portion of the universal picture, and elaborates it.” A great number of body-part metaphors are explained by the fact that body-part functions can be directly observed, that it is easy to interpret body-part names allegorically, and that their system of figurative meanings is complex (Dolgopolov 1973). Body-part expressions have been dealt with by Russian lexicographers. Thus, The Dictionary of Russian Gestures by Dmitrieva et al. (2003) com- prises about 1300 entries. It presents different body-part expressions, repre- senting various human emotional states. For instance, grief may be ex- pressed by the following body-part expressions: , ‘to 140 move the eyebrows’; ( ), ‘to fold one’s hands (in deep melancholy)’; , ‘to shake one’s head’. Chanysheva (1984), studying the representation of gestures in the process of communication, comes to the conclusion that verbal expressions are the most typical expressions in Russian, while in English they are mostly adver- bial and attributive word combinations. Liu & Huang (2006) compared cultural and language traditions in Russia and China, coming to the conclusion that Russians typically differentiate between right and left, symbolizing the Russian attitude to the spatial sys- tem, special landmarks and personal space in general, a distinction that is not obvious in Chinese. According to these analysts, Russians value the right side and despise the left one. To prove this claim, numerous Russian body- part proverbs are presented, e.g. – , – , ‘the right eye itches – to get money, the left eye itches – to cry’. Researching Chinese metaphors based on the sense of vision, Yen’s (2000) thesis utilizes the four figurative categories suggested by Lakoff & Johnson (1999): (a) structural metaphors, (b) ontological metaphors, (c) me- tonymy, and (d) orientational metaphors. He thinks that metaphor in Chinese plays an important role in terms of biology, psychology, and communication. For instance, in the section of “Seeing as Eating Food,” he says that a meta- phorical ‘see’ concept indicating “absorbing food” exists in Chinese. There- fore, speakers of Chinese say 眼饞 ‘to cast a covetous eye on something’ and 看個飽 ‘to see until one has seen enough’. This is because eating is a very important part of Chinese culture, so various ideas can be like various types of food. Body-part studies are also found in mainstream modern linguistics, nota- bly in research on cognitive metaphors. According to an analysis of meta- phoric transfers over three centuries by Smith, Pollio & Pitts (1981), the human body has consistently been a frequent source of metaphors. The rich- ness of body-part metaphors derives from the combination of the universal, cultural, and individual dimensions of our figurative thought. Furthermore, Lakoff & Johnson (1999) argue that the embodiment hypothesis entails that our conceptual structure and linguistic structures are shaped by the peculiari- ties of our perceptual structures. According to them, an embodied philoso- phy shows that the laws of thought are metaphorical, not logical; truth ap- pears to be a metaphorical construction, not an attribute of objective reality. It is obvious that the conceptualization and representation of the body in culturally entrenched codes is an important issue in linguistics. The phrase- ology of any language is anthropocentric. According to Gak (1988), man is egocentric; he puts himself into the centre of the universe and, when com- prehending and categorizing the surrounding world, he compares it with himself. Thus, the names of the upper parts of objects will come from the notion of head, lower ones from the notion of foot, central ones from the 141 notion of heart, and sides are defined by the notion of hand (to the right hand, to the left hand) (Gak 1988: 702). In this paper, we are especially interested in comparative studies of body- part expressions containing words for ‘eye(s)’ and ‘hand(s)’ in Mandarin Chinese, Russian and English. 2.2 Related theories Quite a few studies of human emotion from a linguistic perspective have been published in the last two decades (see e.g. Palmer & Occhi 1999, Atha- nasiadou & Tabakowska 1998, Kövecses 1990, 2000, Niemeier & Dirven 1997, Russell et al. 1995, Wierzbicka 1992, 1997, 1999). A question that comes to light in this research is how the role of the body in emotion or the impact of emotion on the body is conceptualized in different cultures and manifested in different languages. To answer this question requires cross- cultural and cross-linguistic collaboration. Lakoff & Johnson (1980) proposed that concepts based on literal ways of understanding things are extended and repeatedly used in our everyday lan- guage in a figurative way. These are conceptual metaphors. They examine the use of metaphor in English and give a number of examples: LOVE IS MADNESS I’m crazy about her. She drives me out of my mind. He constantly raves about her. He’s gone mad over her. I’m just wild about Harry. I’m insane about her. (Lakoff & Johnson 1980: 49) HAPPY IS UP I’m feeling up. That boosted my spirits. My spirits rose. You’re in high spirits. Thinking about her always gives me a lift. SAD IS DOWN I’m feeling down. I’m depressed. He’s really low these days. I fell into a depression. My spirits sank. (Lakoff & Johnson 1980: 15) 142 In the example LOVE IS MADNESS, the expressions crazy, drives me out of my mind, raves about, etc. are hints that reveal the madness of the affection. They are our life experience expressed in a series of conceptual metaphors such as LOVE IS MADNESS, HAPPY IS UP, SAD IS DOWN. They exist in our day- to-day natural languages. Such metaphors are used not only in our language but in our daily life, and also in our thoughts and actions. Lakoff & Johnson suggest that our conceptual system is metaphorically structured; in other words, “most concepts are partially understood in terms of other concepts” (1980: 56). For example, the concept UP is grounded from the bodily experience that we have bodies and our bodies stand erect. Our concep- tual system is based on simple spatial concepts, such as UP-DOWN, FRONT- BACK, IN-OUT, NEAR-FAR. They further explain that physical experience is “never merely a matter of having a body of a certain sort; rather every experi- ence takes place within a vast background of cultural presuppositions” (1980: 57). The same basic embodied experiences, in which many conceptual meta- phors are grounded, may, however, be defined differently by different cultural beliefs and values (Gibbs 1999). We also draw upon the Sapir-Whorf Hypothesis (Carroll 1977, Whorf 1956). The idea that languages both reflect and shape their speakers’ mental- ities goes back as far as Humboldt’s work and this famous hypothesis. While the Sapir-Whorf Hypothesis was empirically and theoretically challenged by the proponents of linguistic innateness and universality, it has seen a come- back in the works of Wierzbicka (1992, 1997, 1999). Though hardly popular nowadays in its strong version, the weaker claim of the Sapir-Whorf Hy- pothesis, i.e. that language reflects mentality, is very much a part of today’s ethnolinguistic discourse. Our body is essential, not only for living, but also an indispensable source of our language. Body parts are consistently invoked in languages to express a variety of concepts in different cultures. We hope that a comparison be- tween body-part expressions in Chinese, Russian and English will more clearly reveal this fact. It should be noted that the present research focuses on a subset of daily language to study conceptual metaphors, viz. fixed expressions. Fixed ex- pressions are different from creative literary expressions and personal usag- es. They are commonly accepted and frequently used language devices. A fixed expression is traditionally defined as a string of words behaving as a unitary lexical item. Various terms are used to describe fixed expressions, such as freezes, binomials and frozen locutions (Pinker & Birdsong 1979, McCarthy 1990, Landsberg 1995, Moon 1998). According to Moon (1998: 2), who pro- poses a broader approach to fixed expressions, they include metaphors, similes, proverbs, sayings, frozen collocations, grammatically ill-formed collocations and routine formulae. 143 3 Data in Chinese, Russian and English Eyes and hands are integral parts of languages, revealing a new dimension of the mind and presenting thought in action. Although the human body is a po- tentially universal source domain for expressions structuring abstract concepts, cultural groups set up “specific perspectives from which certain aspects of bodily experience or certain parts of the body are viewed as especially salient and meaningful in the understanding of those abstract concepts” (Yu 2003: 13). A group of body-part mappings from source domains to target domains fundamentally arise from the interplay between body and culture. The eyes and hands are external body parts. According to one encyclope- dia, the eye is “the organ of vision and light perception. In humans the eye is of the camera type, with an iris diaphragm and variable focusing, or accom- modation” (Columbia Encyclopedia, 2007). The hand is a “grasping organ at the end of the forelimb of certain vertebrates, exhibiting great mobility and flexibility in the digits and in the whole organ” (Encyclopædia Britannica, 2007). In man, the hand is used mostly for feeling and grasping (Wenig, 2007). We present and discuss the body-part expressions in figures of eyes (Chinese 眼 and 目, Russian ) and hands (Chinese 手, Russian ) in the given languages according to the emotions they express, as well as the conceptual metaphors that are revealed. Table 1 on p. 145 lists emotions that are expressed by eye and hand ex- pressions in these three languages. In this section we will present and discuss the given metaphors and metonyms standing for emotions. The part on met- aphors will follow this order: (1) a presentation of eye and hand for emotion, (2) the metaphors that show physiological effect for emotion, and (3) the metaphors that present body-part movements for emotion. The Chinese vehicle eyes is shared by 目 and 眼. They are sometimes in- terchangeable, as in 兩目昏花 and 兩眼昏花 ‘two-eyes-confuse-flower’, i.e. ‘cannot see clearly due to confused mind’. Sometimes either 目 (e.g. 目不識丁 ‘eye-no know-A’, i.e. ‘totally illiterate’) or 眼 (e.g. 老花眼 ‘old- flower-eye’, i.e. ‘presbyopia; disoriented eyesight, sometimes with the im- plication of being bedazzled’) is the only option. Many vehicles are inter- changeable in English and Russian. Most often eye is interchangeable with face, as in English to look somebody in the eye/face ‘to look directly at someone without fear or shame’ and Russian / ‘to call [someone] a villain to [his/her] eyes/face’. The eye ex- changes places with the ear in English easy on the eye/ear and Russian / ‘not to move the eye/the ear’, i.e. ‘to pay no attention to, to disobey’. Eyes and nose are interchanged in Russian / ‘not to show [one’s] eyes/nose’ i.e. ‘to stay away for a long time, to be absent’, eyes and lips in Russian / - ‘eyes/lips flamed up about something’, i.e. a strong desire for some- 144 Table 1. Emotions expressed in eye and hand expressions in three languages Parts Chinese Russian English Eye Pure emotion: Pure emotion: embar- Pure emotion: happiness, sadness, rassment, sorrow, pleas- love, affection, anger, hate, love, ure, resentment, shame, happiness, fear, shame, dislike, fear, (exaggeration of) anguish shame, shyness, helplessness, surprise, surprise, greed jealousy Physiological effect for emotion: anguish, anger, Physiological ef- Physiological effect insincere grief, malice, fect for emotion: for emotion: admira- indifference, tenderness, admiration, pride, tion, respectability, contentment, fear, sad- desire, sexual inter- surprise, dishonesty, ness, compassion, pas- est adoration, contempt, sion hatred, admonition Hand Pure emotion: Pure emotion: different Physiological ef- sadness, anger emotions, happiness, fect for emotion: pain, sorrow, grief, des- admiration, sur- Physiological effect pair, anger prise, respectability, for emotion: hatred, pride, dishonesty, hostility, admiration, Physiological effect for seriousness, firm- respectability, emotion: merriment, ness, pity, ungrate- thoughtfulness fury, happiness, passion fulness, content- ment thing was aroused’. The interchange of body parts in set expressions is an area deserving of further research. 3.1 Metaphors Our data in three languages can be classified into three groups of metaphors expressing emotions: (1) eye and hand for emotion, (2) eye and hand for physiological effect of emotion, (3) eye and hand for movements, expressing emotions. 3.1.1 Eye and hand for emotion A large group of metaphors comprising the vehicles eye or hand will reveal emotions in the three languages. 145 Eyes (1) Chinese a. 刺目 ‘thorn-eye’, i.e. ‘something attention-catching, or that incurs dislike from others’ b. 盲目不盲心 ‘blind-eye-no-blind-mind’, i.e. ‘physically but not mentally blind’ c. 眼花耳熱 ‘eyes-flower-ears-hot’, i.e. ‘drunk and feeling excited; various kinds of knowledge’ (2) Russian a. ‘[sb] wants to eat with [her/his] eyes’, i.e. an expression of anger, malice b. ‘to throw dust into [sb’s] eyes’, i.e. ‘to boast about imaginary success’ c. , ‘eyes are crying, but the heart is laughing’, about insincere emotions, hypocrisy d. ‘oily eyes’, i.e. an artificially gentle, sugary smile (3) English a. eyesore ‘a thing that is unpleasant to look at’ b. Here’s mud in вour eвe! ‘said by people in a friendly way just be- fore drinking an alcoholic drink together’ c. bedroom eyes ‘seductive-looking eyes or glances which are con- sidered seductive’ Hands (4) Chinese a. 心狠手辣 ‘heart-cruel-hand-spicy’, i.e. ‘callous and cruel; ruth- less; relentless’ b. 手眼通天 ‘hand-eye-reach-sky’, i.e. ‘adept at currying favour with the all-powerful’ c. 眼高手低 ‘eyes-high-hand-low’, i.e. ‘high in aim but low-rate in execution, having high ambition but no real ability; fastidious and demanding but inept’ (5) Russian a. ‘with dirty hands’, i.e. ‘dishonest’ b. ‘to heat [one’s] hands’, i.e. ‘to get profits in a dishon- est way’ c. o- ‘to carry sb in one’s arms’, i.e. ‘to pamper, to give a treat’ (6) English a. a firm hand ‘strong control’ b. cold hands, warm heart: said to someone with cold hands in order to stop them being embarrassed c. come/go cap in hand ‘to ask someone for money or help in a way which makes you feel ashamed’ 146 Different aspects of emotions are expressed with eyes in our three languages. The emotions that are shared by all three are pleasure, shame, and fear. Some examples are listed below. (7) Pleasure a. 順眼 ‘sequence-eyes’, i.e. ‘to like sb or sth’ b. ‘to please [an] eye’, i.e. ‘to please sb’ c. easy on the eye ‘pleasant to look at’ (8) Shame a. 丟人現眼 ‘throw-people-show-eyes’, i.e. ‘to lose face in front of people’ b. ‘to hide one’s eyes’, i.e. ‘to avoid sb’s eyes in or- der to dissemble shame or embarrassment’ c. not look somebody in the eye ‘to be too ashamed to look at sb di- rectly and speak truthfully to him/her’ (9) Fear a. 怵目驚心 ‘scare-eyes-shock-heart’, i.e. ‘startling; shocking’ b. ‘fear has big eyes’, i.e. ‘a frightened per- son sees danger everywhere’ c. shut one’s eyes to ‘to refuse to see’ Many emotions are expressed in this manner only in one of these languages. Expressing jealousy, depreciation, helplessness, surprise, and happiness in this fashion is unique to Chinese. The emotional processes discussed here are supposedly visible. For example, red eyes show jealousy, thus 分外眼紅 ‘divide-outer-eyes-red’, i.e. ‘to envy or admire sb’; staring eyes show sur- prise, therefore 目瞪口呆 ‘eye-stare mouth-silly’, i.e. ‘to stare in bewilder- ment or mute amazement’. In Russian, embarrassment, shame, sorrow, resentment are specially ad- dressed. For example, ‘to cry one’s eyes out’, i.e. ‘to experience utmost sorrow,’ as crying, weeping, tears are manifestation of sorrow; ‘to hide one’s eyes’, i.e. ‘to avoid sb’s eyes in order to dissemble shame or embarrassment’, as humans display shame by drop- ping their eyes, lowering their gaze. In English, desire and sexual interest are expressed with eyes, as in eye something up ‘to look closely at sth that you are interested in’, eye somebody up ‘to look at sb with sexual interest’, and a roving eye ‘sb who is sexually interested in people other than their partner’. The vehicle eyes can express personal qualities, which is a particular characteristic of Russian. For example, courage and indifference are personal qualities, conveyed by Russian eyes. Courage is expressed when men can look into the eyes of danger as ‘to look into the eyes of something’, i.e. ‘to be determined in difficult situations’. On the oth- er hand, not to blink your eye, looking at your interlocutor, means to show 147 nonchalance, as ‘not to blink an eye’, i.e. ‘not to blink one’s eye, demonstrating calmness/indifference’. As for Chinese, the eye functions for arrogance, e.g., 目使頤令 ‘eye-use rear-order’, i.e. ‘arrogant and despising orders’. There is no related expres- sion for such a personal quality in the English data that we have collected so far. The positive emotion of happiness is the most often expressed emotion in the form of eyes. It is shown in the following examples. (10) Happiness a. 眉開眼笑 ‘eyebrow-open-mouth-laugh’, i.e. ‘to beam with joy; a person with a discerning eye; a person of good sense’ b. ‘happy eye’, i.e. ‘a kind eye, having no magical power to injure or harm people by looking at them’ c. there аasn’t a drв eвe in the house: conveys that all the people at a particular place felt very emotional about what they had seen or heard and many of them were crying In terms of hands, some emotions are expressed only in one of these lan- guages. For example, happiness and despair are only found in Russian, as in ‘to break hands’, i.e. ‘to wring one’s hands, expressing grief, misfortune, despair’ and - ‘sb has a light hand’, i.e. ‘somebody brings happiness, luck’. As mentioned above, the way Chinese uses eye to express the emotion of ‘pleasure’ is different from that of Russian and English because the latter two give the viewpoint from the object, but Chinese takes the viewpoint from the speaker. In terms of the object itself, Chinese uses the vehicle hands, for example, 炙手可熱 ‘broil-hand-can-hot’, i.e. ‘of intense political power; burning to the touch’. 3.1.2 Metaphors expressing physiological effect for emotion A wide range of eye and hand expressions in our data are variants of a basic cognitive metaphor PHYSIOLOGICAL EFFECT FOR EMOTION. For example, when a person is upset or in low spirits, his muscles are not well-toned, his brain isn’t alert, and as a result the person doesn’t properly coordinate his body movements, including his hands. He can easily drop something, hence the expressions ‘everything is dropped out of hands’, i.e. ‘work doesn’t go well’, ‘out of hands badly’, i.e. ‘very badly’ show depression, or a low mood. On the other hand, if a person takes control over his body and is determined to do something, he is well coordi- nated and his hand won’t shake (when doing something), as in p - ‘[sb’s] hand would not shake when doing sth’, i.e. ‘[sb] would not be afraid to do something’. 148 Below we give examples in all three languages. Eyes (11) Chinese a. 橫眉豎眼 ‘horizontal-eyebrow-upright-eyes’, i.e. ‘to glare in an- ger’ b. 目瞪口呆 ‘eye-stare mouth-silly’, i.e. ‘be stupefied; stare in be- wilderment or mute amazement’ c. 瞠目結舌 ‘stare-eyes-knot-tongue’, i.e. ‘stare dumbfounded’ (12) Russian a. ‘without batting an eyelid’, i.e. ‘showing no embarrassment’ b. ‘[sb’s] eyes radiated gentle joy’ c. ‘to open [one’s] eyes’ = ‘to stare goggle-eyed at something in astonishment’ (13) English a. not look somebody in the eye/face ‘be too ashamed to look at peo- ple directly and speak truthfully to them’ b. there аasn’t a drв eвe in the house ‘all the people at a particular place felt very emotional about what they had seen or heard and many of them were crying’ Hands (14) Chinese a. 手足無措 ‘hand-foot-no-put’, i.e. ‘to have no idea what to do with one’s hands and feet’ b. 情同手足 ‘love-same-hand-feet’, i.e. ‘[the two] are close like brothers’ c. 手麻腳軟 ‘hand-numb-foot-soft’, i.e. ‘hands feeling numb, feet feeling feeble and weak’ d. 手舞足蹈 ‘hand-dance-feet-dance’, i.e. ‘to dance for joy’ (15) Russian a. ‘to break hands’, i.e. ‘to grieve, to mourn’ b. ‘[sb’s] hands are growing numb’, i.e. ‘[sb] does not want to do anything for grief/want of of energy’ c. ‘to clap [one’s] hands’: sign of amazement, astonishment (16) English a. overplay your hand ‘to spoil your chance of success by saying or doing too much’ b. get out of hand ‘a situation cannot be controlled any more’ 149 Thus, our data demonstrate that the conceptual metaphor PHYSIOLOGICAL EFFECT FOR EMOTION is represented in Mandarin Chinese, Russian and Eng- lish. The physiological effect for emotion is merriment, happiness, passion or fury in Russian hands, and anguish, insincere grief, indifference, merriment, tenderness, contentment, fright, sadness, compassion, sadness, malice, anger or passion in Russian eyes, whereas appreciation, depreciation or admonish- ing occur in Chinese. 3.1.3 Metaphors Expressing Body-Part Movements for Emotion Our data comprise metaphors expressing emotions through the description of eye and hand movements. In the previous group of emotion metaphors, ex- pressing physiological effect of emotions, emotions also go together with or are revealed through movement, but the movement has a physiological basis. There exist various classifications of gestures, or body actions, but most of them include emotive or expressive ones. Devkin, studying the interrelation of emotion and paralanguage, states that “paralanguage means sometimes can help when we lack a certain word and always function when the infor- mation is of emotive character” (1973: 92). The metaphors below presenting body-part movement have symbolic or cultural meaning. Eyes (17) Chinese a. 乾瞪眼 ‘dry-stare-eyes’, i.e. ‘to look on anxiously unable to help’ b. 望眼欲穿 ‘look-eye-want-penetrate’, i.e. ‘to aspire earnestly’ c. 手急眼快 ‘hand-hurry-eyes-fast’, i.e. ‘dexterous, adroit’ (18) Russian a. - ‘to raise [one’s] eyes at somebody proudly and arrogantly’, i.e. ‘to look up proudly and arrogantly’ b. ‘not to know where to direct [one’s] eyes’: expressing embarrassment c. ‘[sb’s] eyes were running in a guilty and cowardly manner’, i.e. ‘[sb’s] eyes were moving from side to side in a guilty and cowardly manner’ (19) English a. not bat an eye ‘not blink an eye’, i.e. ‘to show no shock or sur- prise’ b. look somebody in the eye/eyes ‘to look directly at someone with- out fear or shame’ c. I’m afraid to look her in the eye ‘I feel ashamed’ 150 Hands (20) Chinese a. 手無縛雞之力 ‘hand-no-truss-up-chicken-strength’, i.e. ‘to lack the strength to truss up a chicken; to be feeble; to be physically very weak’ b. 甕中捉鱉 手到擒來 ‘jug-in-catch-turtle’ (‘hand-reach-catch- come’), i.e. ‘catch a turtle in a jar; go after an easy prey’ c. 比手畫腳 ‘compare-hand-draw-feet’, i.e. ‘to gesture with [one’s] hand or feet to indicate meaning; make lively gestures (while talk- ing)’ (21) Russian a. e ‘to wave [one’s] hand sadly’ b. ‘to fold [one’s] hands in deep melancholy’ c. ‘to seize [sb’s] hand in fear’ (22) English a. rub hands with pleasure ‘to experience or display pleased antici- pation, self-satisfaction, or glee’ b. аring one’s hands ‘to clasp and twist or squeeze one’s hands in distress’ c. laв violent hands on one’s self ‘to injure oneself’, i.e. ‘to commit suicide’ 3.2 Metonyms A wide range of different emotions may be presented with eye and hand as metonyms standing for the whole body in the three languages examined. Linguists still argue about whether there is a distinct border between meta- phors and metonyms. The initial process for coining metonymies focuses on a prominent part (the eyes, or hands) of the whole (the person). The expres- sions of emotion yield many metonymies. In Chinese data some expressions can either be metonym or metaphor, e.g., 瞪白眼 ‘stare-white-eyes’, i.e. ‘hate, look down or in anger at someone’. When we consider that the whites of the eyes are part of the eyes, this expression is a metonym. On the other hand, to show the whites is to show the emotion of hate, anger or looking down on someone, and is then a metaphorical expression. In Russian and English particular instances of conceptual metonyms are popular: part for the whole (synecdoche) and symbol for the thing symbolized. Eyes (23) Chinese a. 蜂目豺聲 ‘bee-eyes-jackal-sound’: describes sb wicked b. 眼線 ‘eye-thread’, i.e. ‘informer’ 151 (24) Russian a. ‘angry eyes’, i.e. ‘angry person’ b. ‘frightened eyes’, i.e. ‘frightened person’ c. ‘joyful eyes’, i.e. ‘joyful person’ (25) English a. private eye ‘a person whose job is discovering information about people’ b. anxious eye ‘anxious person’ c. roving eye ‘a person who is sexually attracted to people other than his/her partner’ Hands (26) Chinese a. 扼腕 ‘repress-wrist’, i.e. ‘sorrow or despair’ b. 捶胸頓足 ‘beat-chest-pause-foot’, i.e. ‘grief or deep sorrow’ c. 額手相慶 ‘forehead-hand-each-celebrate’, i.e. ‘formerly, to place hands over forehead in greeting or congratulation’ (27) Russian a. ‘happy hand’, i.e. ‘happy person’ b. ‘masterful hand’, i.e. ‘masterful person’ c. ‘unselfish hand’, i.e. ‘unselfish person’ d. ‘shameless hand’, i.e. ‘shameless person’ (28) English a. leading hand ‘the most experienced person in a factory etc.’ b. an iron hand/fist in a velvet glove ‘someone who seems to be gen- tle but is in fact severe and firm’ c. dab hand ‘someone who is very good at a particular activity’ d. old hand ‘someone who is very experienced and skilled in a par- ticular area of activity’ e. the hand that rocks the cradle (rules the world): said to emphasize that women have a strong influence on events through their chil- dren 4 Discussion Emotion concepts in human languages, as Lakoff (1987) proposed, are known to make use of metaphors and metonymies relating to physiological effects and behavioural reactions. Metaphors and metonymies are fundamen- tal types of cognitive models, both experientially motivated. Lakoff & Turner (1989), Lakoff (1993), and Kövecses (1990, 1995a, 1995b, 2000) have shown that English metaphors and metonymies used in models of emo- tions are motivated by our bodies and physiology. 152 Emotion concepts are expressed through the description of body-part movements. Hands are located at the ends of our upper limbs, with the pri- mary functions of locomotion and grasping objects. Some expressions de- scribe emotions by referring to the bodily movements in reaction to those emotions, e.g., turn one’s hand to an activitв or skill ‘to start doing sth new’ and leading hand ‘the most experienced person in a factory etc.’. These ex- pressions underscore the concurrent, interconnected nature of actions and emotions. As Huang (2002) points out, “One might even venture the claim that the core of an emotion is not simply a psychological state or process, but a readiness to act in a certain way”, the acts have being formulated in the language and expressed with the eyes and hands for bodily actions. The great number of set expressions representing eye and hand movements in these three languages prove that claim. Furthermore, an emotion concept, as Kövecses (2000: 186) has insightful- ly observed, stimulates a complex of social, cognitive, and psychological content, arranged in a more or less stable configuration. It can reveal how different languages and cultures can vary. The purpose of our cross-cultural study was to show differences in emotional life, interpretation of emotions and in metaphoric and metonymic understanding of emotions in different languages. The eye and hand metaphors and metonyms presented above are used to express emotions in Chinese, Russian and English, which is language uni- versality, but the ways they express different emotions differ from language to language, supporting the Sapir-Whorf Hypothesis in its weaker version, that language reflects mentality and culture in a specific way. Metaphors representing emotions, including metaphors expressing physiological effect for emotions, and metaphors expressing body-part movement for emotions, together with metonyms standing for people experiencing emotions, are characteristic for all three languages. If some emotions are expressed with eye and hand metaphors and metonyms present in all three languages, other emotions are expressed by eye and hand only in a particular language. The way eye and hand expressions present emotions in these three lan- guages also differs. For instance, the examples in (7) (in section 3.1.1) show that the Chinese vehicle eyes is speaker-oriented, while Russian and English speak for the object. That is, in terms of eyes, in Chinese, the object is pleas- ant to look at or is great for use because “we people” see it this way, while in Russian and English, the object is pleasant or great in and of itself, it has this quality. On the other hand, the Chinese expression 不順眼 ‘not-sequence-eyes’, i.e. ‘to dislike someone or something’, also expresses that when the object is not pleasant, the speaker takes the responsibility, too; it is not that the object is not good enough, but the speaker has his own judgment. In terms of se- mantic phenomena, the Chinese vehicle eye is used to express the whole person, the insight and viewpoint of this person, which is an instance of me- 153 tonymy. In Russian, by contrast, eye stands for the whole person, experienc- ing emotion, as in ‘timid eyes’; eye expresses the physiological effect of an emotion, in order to stand for the emotion, as in ‘to cry one’s eyes out’, i.e. ‘to experience utmost sorrow’; eye stands for an object, arousing emotions in a person, as in ‘to look into the eyes of something’, i.e. ‘to be determined in difficult situa- tions’, where any danger or difficulty is personified and viewed as having eyes and opposing a human. This is the way matters stand in English. Our data do not support Chanysheva’s hypothesis (1984) that body-part expressions, representing gestures in the process of communication, are mostly verbal in Russian and mostly attributive word combinations in Eng- lish. Our data comprise both verbal and attributive expressions in the two languages compared, cf. Russian (29, 30) and English examples (31, 32). Russian (29) verbal expressions a. ‘not know where to direct [one’s] eyes’ (expressing embarrassment) b. e ‘to wave [one’s] hand sadly’ (30) attributive expressions a. ‘happy hand’, i.e. ‘happy person’ b. ‘fear has big eyes’, i.e. ‘a frightened per- son sees danger everywhere’ English (31) verbal expressions a. shut one’s eвes to sth ‘to refuse to see’ b. not bat an eye ‘to show no shock or surprise’ (32) attributive expressions a. there аasn’t a drв eвe in the house: conveys that all the people at a particular place felt very emotional about what they had seen or heard and many of them were crying b. an iron hand/fist in a velvet glove ‘someone who seems to be gen- tle but is in fact severe and firm’ We also noted that many eye and hand metaphors ‘misuse’ the function of body parts, e.g., eyes are not to see with, but to eat, smell, hear with—that is, an application of synaesthesia on body parts. Thus, in the Russian example ‘[sb] wants to eat with [her/his] eyes’, i.e. an expres- sion of anger, malice, eyes are used as an organ of taste. In the Chinese ex- ample 心狠手辣 ‘heart-cruel-hand-spicy’, i.e. ‘callous and cruel; relentless’, the hands can ‘taste’ peppery hot. It is also interesting to note that in expressions with interchangeable body parts, some body parts, being the organ of different senses, are used inter- 154 changeably, as in / ‘not to move the eye/the ear = not to pay attention to, not to obey’. 5 Conclusion This study presents eye and hand metaphors and metonyms representing emotions in Chinese, Russian and English in the light of the Sapir-Whorf Hypothesis (Carroll 1977, Whorf 1956) and Conceptual metaphor (Lakoff & Johnson 1980). The source domains eyes and hands give abundant mappings in the emo- tion domain in all three languages. The basic model is EYE/HAND STANDS FOR EMOTION. But the models EYE/HAND HAVING A PHYSIOLOGICAL EFFECT STANDS FOR EMOTION and EYE/HAND MOVEMENTS STAND FOR EMOTION are also very popular in these three languages. Besides metaphorical mappings, we have studied metonymical mapping from eye and hand domains to a spe- cific emotion domain, representing a whole person experiencing an emotion. The above mappings are one universality in Chinese, Russian and English, but their language representation differs from language to language, follow- ing the weaker version of Sapir-Whorf Hypothesis that language influences the mentality of the people. Some emotions are expressed with eyes in all three languages, such as pleasure, shame, and fear, in which, however, the Chinese vehicle eyes is speaker-oriented, a subjective point of view, while Russian and English are speaking for the object, an objective point of view. In terms of the object itself, Chinese can use the vehicle hands. The positive emotion of happiness is the most often expressed emotion in the form of eyes. In the form of hands, not many emotions are shown in all three languages. Many emotions are expressed only in one of these lan- guages. For example, happiness and despair are only found in Russian. Only English hands express the emotion of satisfaction. Eye and hand, being very important body parts, yield a great deal of mate- rial for further research, both in the sphere of cognitive linguistics and in the cross-linguistic and cross-cultural study of set expressions. Several issues touched upon in this paper, e.g. synaesthetic transfers in body-part expres- sions, the syntactic structure of eye and hand expressions, and metaphor- metonymic switches of body-part expressions, offer new topics for further investigation. 155 Acknowledgements We would like to thank the anonymous reviewers for their valuable com- ments on this article. This study was partially supported by the research grant from the National Science Council (NSC 96-2411-H-006-022, Cogni- tive Foundations of Embodiment Expressions in Mandarin, German and English). References Aitchison, Jean. 1994. Words in the Mind: An introduction to the mental lexicon. 2nd ed. Cambridge, MA: Blackwell. Athanasiadou, Angeliki, & Elżbieta Tabakowska, (eds). 1998. Speaking of Emo- tions: Conceptualization and expression. Berlin: Mouton de Gruyter. Bally, Charles. 1951. Traité de Stylistique Française [Treatise on French Stylistics]. 3rd ed. Paris: Klincksieck. Carroll, John Bissell (ed.). 1977 [1956]. Language, Thought, and Reality: Select- ed writings of Benjamin Lee Whorf. Cambridge, Mass: MIT Press. Chanysheva, Zulfira. 1984. Vzaimodeĭstvie iazykovykh i neiazykovykh faktorov v protsesse rechevogo obshcheniia [Interaction of Linguistic and Non-linguistic Factors in the Process of Speech Communication]. Ufa: Bashkiria State Univer- sity Press. The Columbia Encyclopedia, 6th ed. 2007. New York: Columbia University Press, 2001–07. On-line at: www.bartleby.com/65/ (retrieved on 10 March 2007). Devkin, Valentin. 1973. ‘Nekotorye osobennosti “kineticheskogo” koda i ego vzai- modeĭstviia c verbal’nym’ [Some Characteristics of “Kinetic” Code and its In- teraction with Verbal Code], in Kosjakin, A. N. (ed.), Voprosy stroia nemetskoĭ rechi [On the Structure of German Speech]. Vladimir: Vladimir State University Press, 92–100. Dmitrieva, Ludmila, Larisa Klokova & Valentina Pavlova (eds). 2003. Slovar’ iazyka zhestov [The Dictionary of Russian Gestures]. Moscow: AST Press. Dolgopolov, Yury. 1973. Sopostavitel’nвĭ analiz somaticheskoĭ frazeologii (na ma- teriale russkogo, angliĭskogo i nemetskogo iazykov) [A Comparative Analysis of Somatic Phraseology in Russian, English and German]. Doctoral Dissertation. Moscow: Moscow Pedagogical University. Encyclopædia Britannica. 2007. Encyclopædia Britannica, Inc. On-line at: http://www.britannica.com/eb/article-9039122 (retrieved on 12 March 2007) Gak, Vladimir. 1988. ‘Metafora: universal’noe i spetsificheskoe’ [Metaphor: Uni- versal and specific], in Telia, V. N. (ed.), Metafora v iazyke i tekste [Metaphor in Language and Text]. Мoscow: Science Press, 11–26. Gibbs, Raymond W. Jr. 1999. ‘Taking Metaphor out of our Heads and Putting it into the Cultural World,’ in Gibbs, Raymond W. Jr. & Gerard Steen (eds), Metaphor in Cognitive Linguistics. Amsterdam: John Benjamins, 145–166. Heine, Bernd. 1997. Cognitive Foundations of Grammar. Oxford: Oxford Universi- ty Press. Heine, Bernd, Ulrike Claudi & Friederike Hünnemeyer. 1991. ‘From Cognition to Grammar: Evidence from African languages’, in Traugott, Elizabeth Closs and Bernd Heine (eds), Approaches to Grammaticalization, 1. Amsterdam: John Benjamins, 149–188. 156 Hollenbach, Barbara E. 1995. ‘Semantic and Syntactic Extensions of Body-part Terms in Mixtecan: The case of “face” and “foot”’, in International Journal of American Linguistics 61(2): 168–190. Hsieh, Shelley Ching-yu (謝菁玉). 2009. Embodiment in Languages (1): Human, animal and plant expressions. Taipei: Bookman. — 2008. ‘Das Erscheinungsbild von Pflanzen und Tieren in deutschen Redens- arten’ [The representation of plants and animals in German dialects], in Deutsch-taiwanische Hefte 14: 103–121. Huang, Shuanfan (黃宣範). 2002. ‘The Pragmatics of Focus in Tsou and Seediq’, in Language and Linguistics, 3(4): 665–694. Kövecses, Zoltán. 2000. Metaphor and Emotion: Language, culture, and body in human feeling. Cambridge: Cambridge University Press. — 1995a. ‘Metaphor and the Folk Understanding of Anger’, in Russell et al. 1995: 49–71. — 1995b. ‘Anger: Its language, conceptualization, and physiology in the light of crosscultural evidence’, in Taylor, John R. & Robert E. MacLaury (eds), Lan- guage and the Cognitive Construal of the World. Berlin: Mouton de Gruyter, 181–196. — 1990. Emotion Concepts. New York: Springer-Verlag. Lakoff, George. 1993. ‘The Contemporary Theory of Metaphor,’ in Ortony, Andrew (ed.), Metaphor and Thought. 2nd ed. Cambridge: Cambridge University Press, 202–251. — 1987. Women, Fire, and Dangerous Things: What categories reveal about the mind. Chicago: University of Chicago Press. Lakoff, George & Mark Johnson. 1999. Philosophy in the Flesh: The embodied mind and its challenge to Western thought. New York: Basic Books. –– 1980. Metaphors We Live By. Chicago: University of Chicago Press. Lakoff, George & Mark Turner 1989. More than Cool Reason: A field guide to poetic metaphor. Chicago: University of Chicago Press. Landsberg, Marge. 1995. ‘Semantic Constraints on Phonologically Independent Freezes’, in Landsberg, Marge E. (ed.), Syntactic Iconicity and Linguistic Freezes: The human dimension. Berlin: Mouton de Gruyter, 65–78. Lin, Qing-yuan (林清淵). 2003. The Metaphors and Metonymies of Body Parts in Taiwanese Southern Min. M.A. Thesis. Minxiong (Taiwan): National Chung- Cheng University. Liu, Guang-zhun & Su-hua Huang (劉光准 & 黃蘇華). 2006. The Customs of Lan- guages in Russian and Chinese. Beijing: Foreign Language Teaching and Re- search Press. McCarthy, Michael. 1990. Vocabulary. Oxford: Oxford University Press. Moon, Rosamund. 1998. Fixed Expressions and Idioms in English. Oxford: Claren- don Press. Niemeier, Susanne & René Dirven (eds). 1997. The Language of Emotions: Concep- tualization, expression, and theoretical foundation. Amsterdam: John Benjamins. Palmer, Gary B. & Debra J. Occhi (eds). 1999. Languages of Sentiment: Cultural constructions of emotional substrates. Amsterdam: John Benjamins. Pinker, Steven & David Birdsong. 1979. ‘Speakers’ Sensitivity to Rules of Frozen Word Order’, in Journal of Verbal Learning and Verbal Behavior 18: 497–508. Russell, Jacob A., José-Miguel Fernández-Dols, Anthony S.R. Manstead & Jane C. Wellenkamp (eds). 1995. Everyday Conceptions of Emotion. Dordrecht: Kluwer. 157 Smith, Michael K., Howard R Pollio & Michael K. Pitts. 1981. ‘Metaphor as Intel- lectual History: Conceptual categories underlying figurative usage in American English from 1675 to 1975’, in Linguistics 19: 911–935. Sweetser, Eve. 1990. From Etymology to Pragmatics. Metaphorical and Cultural Aspects of Semantic Structure. Cambridge: Cambridge University Press. Tsai, Li-chung (蔡立中). 1994. The Metaphor of Body-parts in Chinese. M.A. The- sis. Hsinchu City (Taiwan): National Tsing-Hua University. Tsao, Feng-fu, Li-chung Tsai & Hsiu-ying Liu (曹逢甫、蔡立中、劉秀瑩). 2001. Shengtiyupiyu: yuyanyurenzhdeshouyaojiemian [Body and Metaphor: The first interface of language and cognition]. Taipei: Crane. Wenig, J. 2007. Encyclopedia Americana. On-line at: http://ea.grolier.com/cgi- bin/article?assetid=0192750-00 (retrieved on 15 March 2007). Whorf, Benjamin. 1956. Language, Thought and Reality. Selected Writings (ed. John B. Carroll). Cambridge, Mass: MIT Press. Wierzbicka, Anna. 1999. Emotions across Languages and Cultures. Cambridge: Cambridge University Press. –– 1997. Understanding Cultures through their Key Words. Oxford: Oxford Univer- sity Press. –– 1992. Semantics, Culture and Cognition. Universal Human Concepts in Culture- Specific Configurations. Oxford: Oxford University Press. Yen, Chao-tai (顏兆泰). 2000. A Semantics Study of Chinese Metaphors Derived from Visual Perception. MA Thesis. Taipei: Fu-Jen Catholic University. Yerofeeva, Irina & Adelia R. Shamsutdinova. 2006. ‘Lingvisticheskie sredstva soz- daniia portretnoĭ kharakteristiki v istorii russkogo iazyka’ [Linguistic Means of Creating Literary Characters in the History of Russian], in III Mezhdunarodnye Boduėnskie chteniia (Kaгan’, 23–25 maia 2006 g.). Tom 1 [Proceedings of the Third International Symposium in Honor of Baudouin de Courtenay. Vol. 1]. Kazan: Kazan State University, 76–78. Yu, Ning 於寧 . 2003. ‘The Bodily Dimension of Meaning in Chinese: What do we do and mean with “hands”?’, in Casad, Eugene H. & Gary B. Palmer (eds), Cognitive Linguistics and Non-Indo-European Languages. Berlin: Mouton de Gruyter, 337–362. –– 1998. The Contemporary Theory of Metaphor: A perspective from Chinese. Am- sterdam: John Benjamins. 158 Anna Kryvenko Conventional Animal Metaphors and Cultural Scenarios in the English and Ukrainian Sports- Related Lexicon Abstract. This study explores the nature of conventional sports-related metaphors represented by lexemes, collocations and idioms with an animal component in Eng- lish and Ukrainian and accounts for extralinguistic factors such as sports practices and culturally bound interaction between humans and animals, which are linked to both epistemic and ontological cognition, as well as intralinguistic features, includ- ing the inner form, semantic change, borrowing, semantic structure, and paradigmat- ic relations of the items under analysis. The evaluation of the data is largely based on the argument that these qualitative aspects reveal the degree of motivation and imagery and therefore explain the cross-linguistic salience and co-occurrence of both the lexical concepts and the cognitive models underlying them. The results of the research are interpreted with regard to distinctive features in the categorization and conceptualization of the world, and a typology of corresponding source and target domains is offered. A close case study of the hunting scenario represented by English and Ukrainian lexemes, collocations and idioms shows that although based on the conceptualization of similar basic experiences, related linguistic expressions may highlight different components of the scenario. Divergence in figurative mean- ings is imposed by both linguistic and cultural factors. Keywords: animal name, conventional metaphor, contrastive analysis, intralinguis- tic knowledge, extralinguistic knowledge, hunting scenario 1 Introduction One way in which cross-linguistic studies of metaphor have taken advantage of the rich potential of modern linguistics is by exploring cognitive catego- ries which are “concealed by purely linguistic phenomena, on the one hand, and areas of extralinguistic knowledge on the other” (Fillmore 1985; Lakoff & Johnson 1980; Lakoff 1987; Langacker 1987; Baranov & Dobrovol’skij 1996: 410). The present research aims to decode and contrast linguistic and conceptu- al meanings of conventional English and Ukrainian animal metaphors related to sports as representations of cultural data on the one hand, and cognitive structures on the other. The relatively compact area of vocabulary chosen for 159 this contrast allows one to account for systemic patterns in lexicalization and conceptualization of man–animal relations in this particular segment of hu- man experience. Although researchers of conceptual metaphor theory often speak of con- ventional metaphors as being entrenched in the minds of individual speakers, the social nature of conventional figurative language should not be dispar- aged (Svanlund 2007: 50). Based on the understanding of conventionality as a common communal coordination device for a recurrent problem (Lewis 1969; Clark 1996; Croft 2000), in this case communicating a meaning that is both linguistic and cultural, thematic clusters of conventional metaphors are viewed in this paper as discrete representations of conceptualized world knowledge constrained by language competence and social behavior. Furthermore, conventional metaphors contribute to the ontologization (Baranov & Dobrovol’skij 1996: 410) or registration of cultural knowledge in natural languages, since they can be understood and used by those mem- bers of language communities who do not have the personal experience re- ferred to in conventional metaphors as well as those who do (cf. the dichot- omy between general and specific knowledge, as discussed by Schank and Abelson (1977: 37)). Conventional metaphors also provide powerful data for contrastive stud- ies when used in cognitive modeling of representations of cultural knowledge. In this paper, English and Ukrainian sports terms that explicitly name or implicitly refer to animals will be examined in two dimensions. First, conventional animal metaphors from the source domain SPORTS pro- jected onto various target domains are compared in English and Ukrainian. Second, lexical representations of the source domain ANIMAL mapped onto the target domain SPORTS are contrasted in the languages concerned. The research procedures also include the identification of “alive and, for the most part, buried” root analogies (Goatly 1997: 45) that underlie these conven- tional animal metaphors and the reconstruction of cultural scenarios (related to scripts1 as specified by Schank & Abelson 1977: 38–46), which (in Eng- lish and Ukrainian, respectively) represent typical sequences of events in common animal-related sports. A semantic-role analysis is employed to ex- plore relations between humans and animals in a hunting scenario. To avoid possible vagueness, it should also be mentioned that the poly- semantic term animal is used in this paper to denote not merely mammals 1 Although the terms script and scenario are often used synonymously in cognitive linguistics, and more broadly in cognitive science, our choice of the term cultural scenario is determined by the need to emphasize the procedural nature of the tradi- tional knowledge under study and should not be identified with the term cultural script offered by Wierzbicka (1994) as a research technique specifically developed for abstract concepts. 160 but also members of other zoological classes. The polysemantic term sport(s) refers here mainly to an individual or group activity pursued for exercise or pleasure, involving the testing of physical or mental capabilities, often in the form of a competition. 2 Animal-related sports and their metaphoric representations Common animal-related sports include various forms of racing, riding, fighting and hunting and among the most common animals that participate in sporting events these days are dogs, horses, and bulls. Animal-related sports represent the prototypical features of the category sport(s) to varying de- grees. According to Rosch (1975: 232), polo, at rank 18, is a better example of sports in comparison to fishing, horseback riding, horse racing, and hunt- ing, rated between 41st and 52nd, respectively. Some of these sports are pursued primarily as a pastime (polo, dressage), whereas others are or used to be occupations (hunting, rodeo). Some of them are recognized Olympic sports (equestrian sports, polo), whereas others are practiced illegally or are out of practice these days (cock fighting, dog fighting, badger baiting, or pigeon shooting). The latter are mostly blood sports, where killing animals is a goal. Animal-related sports also vary in levels of participation by humans. Some take place solely between animals (dog racing, cockfighting), while others use animals either as assistance (po- lo) or as targets (bullfighting) (‘Animals in Sport’ 2008). The terminology of animal-related sports and its figurative usage are rep- resented unequally in English and Ukrainian. Some of these sports are lim- ited to English-speaking and some other countries and are not practiced in Ukraine. However, even English terms for such sports as polo or rodeo, popular in the UK and the USA, respectively, do not seem to contribute to the conventional stock of animal metaphors, according to our data. On the other hand, some old-fashioned or illegal sports like bullfighting or cock- fighting preserve their imprints in both English and Ukrainian. This suggests that conventional animal metaphors should be explored both synchronically and diachronically in their relation to the language structure and the material culture of language communities under study. Cockfighting is used in English as a vehicle to discuss physical or verbal STRUGGLE (a battle royal; to pit against; to spar) as well as describe accom- panying emotions, psychological states or undesired qualities, namely AN- GER (to get your hackles up), DEPRESSION (crestfallen) and COWARDICE (show the white feather). The Ukrainian near-equivalents for the abovemen- tioned metaphors do not allude to animals, although some of them are quite similar in their imagery: crestfallen – , literally ‘spirit fall- en’ (DOWN IS BAD), show the white feather – ‘to 161 show light spirit’ (cf. faint-hearted in English). However, ‘cockfight’ can be a political metaphor of a verbal or even physical struggle with an element of pretence or meaninglessness, these days often in refer- ence to members of opposing parties in the Verkhovna Rada (the Ukrainian Parliament) or in the government. Metaphors that allude to dog fighting in English, in addition to STRUG- GLE (a pit bull; to dogfight – ‘to bite each other like dogs’), also include the concept of CONTROL (top dog, underdog), which is not registered in Ukrainian. Bullfighting is associated with EMOTIONS (to see red or like a red rag to a bull – ), CONTROL (to take the bull by the horns – ); and TESTING (moment of truth – ). The complete correlation between these English and Ukrainian metaphors points to their common idiomatic origins, e.g. moment of truth, which is a translation loan of the Spanish el momento de la verdad, the final sword- thrust in a bull-fight. Known to several European languages (cf. Swedish ta tjuren vid hornen, German den Stier (Bock) bei den Hörnern packen (fas- sen), French prendre le taureau par les cornes), the idiom to take the bull by the horns is believed to be motivated either by bullfighting or by agricultural practices. Various forms of horse riding, especially horse races, are among the most common sources of sports-related animal metaphors in both English and Ukrainian. The target domains include not only a more general COMPE- TITION for prizes and rivalry in POLITICS or BUSINESS (Deignan 2005: 28– 29), where a special location or movement is mapped onto career PROGRESS (break out of the pack, down the stretch, left at the post), but also MUSIC (disc jockey – - ), TIME (at the drop of a hat, down to the wire), CONTROL (hold the reins – ), ABILITY (handicap – ), EMOTIONS (hold your horses – ‘careful on the turns’) and so on. For example, horse racing is particularly closely asso- ciated with an election campaign, where the roles of politicians and their actions are described in terms of race participants: a dark horse – ; race for power – . The minimum for getting into the Verkhovna Rada (the Ukrainian Parliament) is described by an allusion to the steeplechase: 3% ‘є ‘to clear the 3% hurdle’, because only those parties that gain more than 3% of the votes in an election can seat MPs. П , literally ‘to take for a ride in a vehicle drawn by black horses’ means ‘a negative vote or veto’ and alludes to the blackball as a method of voting, because the verb can refer to both riding horses and rolling round objects (e.g. balls) and black horses are associated with the color of the ball. An unsuccessful candidate in an election can be dubbed as an also-ran in English, which is a horse racing term, and ‘outsider’ in Ukrainian, which is an English semantic loan and a 162 general sports term primarily denoting a low ranked team or sportsman; in this case the allusion to horse racing in Ukrainian is vague. Not surprisingly, hunting and fishing are the most prolific sources of sports-related animal metaphors both in English and in Ukrainian. Whether or not hunting as a means of obtaining food was a dominant influence on human evolution and cultural development, as opposed to gathering vegeta- tion or scavenging (Petersen 2008: 811), hunting analogies are well en- trenched in our collective cognition and it does not take individual experi- ence to adequately understand and use hunting metaphors, since they are highly conventionalized and part of the traditional environmental knowledge of different language communities. 3 Metaphor networks and cultural scenarios We argue that conventional hunting metaphors, which are realized in lex- emes, collocations and idioms, are not isolated. They lend themselves to natural inference rules and form a conceptual network which accesses a hunting scenario. In particular, this approach is opposed to the traditional understanding of phraseology within which idioms are analyzed as unique and semantically arbitrary phrasal units. In fact, certain laws of logic and human experience govern idioms, on the one hand, and linguistic rules regu- late their formation, on the other (Baranov & Dobrovol’skij 1996: 412). The hunting scenario is a particular type of action scenario, i.e. an ideal- ized model of action, or the Transitive Scenario in the terminology of Pan- ther & Thornburg (2002: 285). The scenario is used in this article as tertium comparationis, since it is oriented towards an algorithmic description of ACTION, and in broader terms “towards such types of knowledge which are procedural in nature” (Baranov & Dobrovol’skij 1996: 416). In this paper, knowledge is understood as a consequence of the procedures peculiar to the functioning of an object or typical of one action or another and specific for a particular cultural environment. A contrastive analysis shows that each slot in the hunting scenario is rep- resented by conventional metaphors in English and Ukrainian (see Table 1, next page). However, this representation is not coherent in either language. Unlike the first and the last slots, metaphors in the other five represent three viewpoints, or vantage points, that of the hunter, that of the hunting dogs and that of the hunted animal. Moreover, the differentiation between the first two vantage points can be implicit or merge into one view, but the dog’s view is clearly opposed to that of the prey: to be on the track of sth – to cover up one’s tracks; to be at fault – to vanish without a trace; to set a trap – to be trapped in both languages. The last three slots also differentiate between success and failure of both hunters and the hunted: to be on a hot trail – to 163 Table 1. A hunting scenario in English and Ukrainian metaphors  open season on something, Setting: English Ukrainian time ;  (happy) hunting ground in season, out of season; location Roles: headhunter, fortune hunter, agent, rock hunter; ( , ); patient, fair game, quarry, prey; ; instrument hound, sleuthhound, bird dog Prey detection keep your ear to the ground, ; sniff sth out, get wind of, cast ; about Raise beat the bushes; break cover ; ; Pursuit on a hot trail, at fault; turn tail, ; vanish without a trace ; ; Killing or cap- in at the death (kill), run to ; ; turing ground, have at bay, shut your ; trap, bag sth, stop sth dead in its tracks come away empty-handed Division of eat humble pie , ; spoils be on a cold trail or to be at fault; to be on the right track – to be on the wrong track; to keep track of sb / sth – to lose track of sb / sth; to bag sth – to come away empty-handed, ‘to divide up the bearskin before the bear is shot’. The level of detail in the slots “pur- suit”, and “killing or capturing” is higher than in other slots. This, as well as the fact that a number of metaphors that fall into these slots are synonymous, e.g. close on the heels of sth, in hot pursuit of sth; on sb’s tail, highlights them as peak activities in a hunting scenario as actualized in both languages. Despite numerous techniques and styles of hunting, a unified hunting sce- nario that appears through conventional metaphors emphasizes two types of relations between its participants: a causal relation between the hunter and the hunted, where Agents greatly affect Patients, and a relation between hunting humans and hunting animals which can be viewed in terms of Agents and Instruments. 164 In fact, the very term hunter – is polysemantic. In addition to its main meaning ‘a person who seeks out and kills or captures game’ in both languages, and its figurative meaning ‘one who searches or pursues sth’, in English it can also denote a wild animal that feeds on other animals, i.e. a predator, as well as a dog or a horse used for hunting. In Ukrainian we would denote these meanings as ‘predator’, ‘hunting dog’, and ‘hunt on horseback’, respectively, but we would not use ‘hunter’ in any of these cases. The reason is that the Ukrainian lexeme ‘hunter’ is more anthropocentric. Figuratively, it can also refer to a smart and sharp-witted man, which is motivated by its inner form. М /myslyvets’/ may be traced all the way back to Proto-Slavic mвsl ‘thought’, which clearly points to MENTAL DOMAIN. Apparently the ancient Slavs treated hunting as a “thoughtful activity” requiring agility of mind. The form has cognates in some other Slavic languages in the region, namely Belarusian, Polish, Czech and Slovak, but not in Russian. The Russian ‘hunter’ is derived from , which means both ‘hunting’ and ‘desire’ and activates EMOTIONAL DOMAIN. The English lex- eme hunter stems from OE hunta, akin to the Gothic ablaut form hinþan ‘to seize, to capture’. Its motivational base activates PHYSICAL DOMAIN, namely ACTIVITY. In fact, a similar motivation base underlies ‘catcher’, an- other Ukrainian word for hunter, common at the time of Kyiv Rus’ but obso- lete these days. To complete the picture, it should be pointed out that the Ukrainian verb ‘to hunt’ stems from ‘field’, which accesses LOCATION, common for hunting in Ukraine. The nature of the conceptual links between hunting humans and hunting animals in a hunting scenario is somewhat uncertain, and its interpretation depends on how a conceptualizer views this relation. From an anthropocen- tric perspective, specially trained animals (commonly dogs, horses or fal- cons) are animate Instruments used by humans in the pursuit of wild ani- mals, and the part of their definition ‘used for hunting’ suggests this reading. However, in Ukrainian there is no fixed generic term to denote a horse used for hunting. Г ‘hunter’ is a transliterated loan from English referring to a particular breed selected in England and Ireland for horseback riding. The collocation ‘to hunt on horseback’ represents the horse as a vehicle, where a part of the animal, i.e. horseback, stands for the whole, and emphasizes in this case the manner of action. Unlike horses, canines and birds of prey perform goal-oriented actions in hunting and can act independently of humans. In the past, dogs had hunted on their own account and man has done no more than correct the dog’s instinctive style of hunting, molding it to the convenience of a collaboration. In order to discover the ex- 165 tremely cautious animal [prey] man resorted to the detective instinct of anoth- er animal [dog]; he asked for its help. (Ortega y Gasset 1972: 88) In fact, helping can be viewed as a subtype of causal relations (Koenig et al. 2007: 175). Thus, by helping the hunter to cause a change of both location and eventually the state in Patient, i.e. the game, dogs perform an instrumental role. The relation between the hunting human and the hunting dog can be re- garded as a metonymic extension, if we assume that “Instruments are part of some sufficiently general human Agent model” (Panther & Thornburg 2002: 298), or as a metaphoric extension, if Instruments act as Agents, in particular if they occur as the subjects in transitive sentences (ibid.). Further analysis of hunting terms mapped onto other domains shows that metaphoric extension is highly productive. Overall, the mapping of the hunting scenario onto other domains is a sampling and the degree of its completeness depends on the number of ac- cess points that connect the source and the target. The greatest similarities are registered between hunting game and detecting criminals in both Eng- lish and Ukrainian, since both activities are highly dynamic and both are structured by the PATH image-schema (Lakoff 1987) or more specifically PATH-TO-GOAL. Policemen, or metaphorically, sleuthhounds or bloodhounds – ‘hounds’, literally ‘searchers’ (about both detectives and dogs), investigate (from vestige ‘track’) – (from ‘track’) a crime. Like hunters they ‘set a trap’ for criminals, but like dogs they hound or bird-dog them, are on the track of them, tail them, are hard on their heels and eventually hunt them down. Criminals, like game, try to throw the police off the track / scent, lie low, when flushed out or smoked out they break cover, yet they are often doomed. If Patients in the hunting scenario are mistreated and do not deserve suf- fering, or — figuratively — criticism, they are depicted as victims, i.e. prey or quarry, though to various degrees in both languages, and hunting is no longer a fair game: open season on sb (cf. ‘witch- hunt’); ‘bait (with dogs)’ – to badger. A person in a defenseless position is a sitting duck or a clay pigeon, someone doomed to death or fail- ure is a dead duck (cf. lame duck ‘sb disabled or ineffectual’; in Ukrainian ‘lame duck’ is a diminutive and a term of endearment for a lame person, often a woman, which is based on a folk image in a popu- lar Ukranian fairy-tale). As opposed to Patients, a purely instrumental role is ascribed to animals (and metaphorically people) used to beguile or lead someone into danger: decoв, stool pigeon, lure, . Searching for information differs from actually detecting criminals or pursuing the innocent, though it can be the preliminary stage of the former and part of the police hunting scenario. Searching for information varies in 166 dynamism, humanness and affectedness, depending on Patients, e.g. celebri- ties or news. Nevertheless, reporters as well as spies are also identified with hounds: they dog celebrities, have a nose for news and sniff out facts, ferret out or get wind of secrets; ‘sniff out’, ‘smell out’, ‘nose out’, ‘rummage’. Most of these lex- emes and idioms metaphorically point to the nose as a tool for gaining in- formation and thus instantiate a conceptual metaphor SMELLING IS KNOW- ING, as opposed to the well-established SEEING IS KNOWING. A recruiter’s search for personnel, or modern ‘headhunting’, is quite humane and it commonly benefits the hunted as well as the headhunters, as opposed to the savage practice of cutting off and pre- serving the heads of one’s enemies as trophies. Unlike in English, in Ukrainian a hunting scenario is mapped onto the rit- ual of matchmaking and engagement. In a traditional Ukrainian engage- ment ritual, a prospective groom invites matchmakers by saying: “Saddle your horses, call your hounds, and let’s go hunting to an open field. I saw a marten in a tree there: the marten is in the tree and the girl is in the tower” (Zhajvoronok 2006: 320–321). Later, matchmakers enter the house of the bride-to-be’s parents or kinsmen with the words: “Our game ran into your house and is hiding here. We are going to catch our marten here” (ibid.). The correlation between a bride and ‘marten’ can be viewed as both a metaphoric and metonymic extension. On the one hand, the resem- blance between a bride and a marten is rooted in erotic symbolism. On the other hand, marten pelts were used as currency and at that time / ‘marten’ denoted a currency unit in the Slavic world (cf. the Croatian national currency, the kuna). The symbolic hunting of a bride replaced a more ancient ritual of stealing a girl from a clan and marrying her without her relatives’ approval. Later it was transformed into a ritual of buying a bride from her kinsmen and paying ‘wedding marten’, i.e. a bride price. In English, however, matchmaking can be viewed in terms of a fishing scenario. In particular, someone well worth marrying can be referred to as a keeper or a good catch and if you hook your partner you win his or her fa- vor. Someone can be hooked on sb, here ‘be in in love with sb’, or hooked ‘married,’ respectively. A somewhat similar mapping also takes place in Ukrainian: , literally ‘to cast the line’ or figuratively ‘to put out a feeler’, ‘a fishing ground’ (cf. happy hunting ground) or ‘a place where there is plenty of what you are looking for’, ‘a good catch’, both literally and figuratively, ‘to hook (a partner)’. On a broader scale, these lexemes, collocations and idioms are commonly used to metaphorically describe many other forms of human activities and relations in both languages. 167 4 Two root animal analogies in non-animal-related sports It has been noted that even those sports which do not involve animals also conventionally use animal imagery in their terminology, both in English and in Ukrainian. We argue that animal references in the terminology of non- animal-related sports are not casual. In fact, they are well structured and they also instantiate a conceptualization and categorization of the world around us. Moreover, they are projections of the interaction between humans and animals from other areas of life. The two root analogies (Goatly 1997: 48–49) that underlie transfers from the source domain ANIMAL to the target domain SPORTS are: 1. HUMANS = ANIMALS, and 2. OBJECTS = ANIMALS The grounds for the first transfer include similarities in traits, motion, or appearances between humans engaged in sports and animals. On a broad scale, lexicalized animal metaphors that fall under root analogy (1) are used as terms to denote mostly holds (Boston crab), throws (flying mare), blocks ( ‘hedgehog’; , , , all terms for ‘stork’), stands ( ‘swallow’, ‘bear’, ‘crane’, ‘frog [diminuitive]’, ‘dog [diminuitive]’, ‘spider [diminuitive]’), spins (camel spin, ‘crocodile’), jumps (butterfly jump, ‘goat [diminuitive]’), dives ( ‘swallow or swan dive’), strokes (butterfly – ‘dolphin’, doggy paddle – - ‘dog- like’), punches and kicks (rabbit punch, ‘colt’, ‘male pike’, ‘hawk’, ‘goat’) in martial arts, wrestling, boxing, gymnastics, figure skating, swimming, and diving. There are a few instances of animal names figuratively referring to partic- ipants of sports events: sportsmen (puck hog ‘a hockey player who is reluc- tant to pass the puck to other members of his or her team’, ducks on a pond ‘runners in scoring position [baseball]’, southpaw ‘a left-hander [baseball, boxing]’, ‘falcon’, ‘hawk’ [both denoting mastery ranks in hopak ]), fans (puck bunny ‘a young female hockey fan, usually more inter- 1 ested in the players than the game’), and referees (zebra). The names of sports clubs, especially in the USA, are also very rich in an- imal imagery, for instance, in baseball (Baltimore Orioles), American foot- ball (Miami Dolphins), ice hockey (Phoenix Coyotes), basketball (Chicago Bulls), or running (Hoboken Harriers). We claim that these are imprints of magic beliefs in the transference of animal power or skills to humans via names by contiguity (cf. names of Native American Indian clans and socie- 1 The Ukrainian Cossack martial art, first described in the 15th century, that has survived over the centuries as a folk dance but is being rediscovered as a sport/ martial art these days. 168 ties), which points to their metonymical nature. In Ukraine, so far there is only one registered sports club with an animal component in its name, the basketball team Ч ‘Cherkassy Monkeys’. The second root analogy (2) underlies mostly terms for sports equipment, whether still or in motion, and scores in gymnastics, shooting and archery, golf, baseball, badminton, American football and some others. Examples include ‘vaulting horse’ or ‘pommel horse’ and – buck ‘a type of vaulting horse’. The phenomenon of meronymy is observed in the following patterns: part of object = part of animal: bull’s eвe; the whole object = part of animal: flippers or fins – , a pigskin ‘a football’ (which was formerly made of pigskin), a horsehide ‘a baseball’ (hasn’t been made of horsehide since 1985); Ukrainian ‘slingshot’ (motivated by ‘horn’); part of object = the whole animal: ‘dog’ for trigger. A hunting scenario maps onto other sports in English but not in Ukrainian. In particular, targets are associated with animals: bull’s eвe (cf. ‘apple’ in Ukrainian), clay pigeon (cf. ‘small plate’); a struck ball resembles a flying bird: (in badminton) a shuttlecock (cf. ‘shuttlecock’ but also ‘flounce’); (in golf) a birdie, an eagle/double eagle, a duck hook ‘a short tee shot that travels hard and fast into the left rough’ (cf. a worm- burner ‘a low golf shot’); (in baseball) a dying quail ‘a fly ball that suddenly drops just beyond the infielders’; (in American football) a wounded duck ‘a wobbly pass’. A baseball is also associated with small and quick mammals: a rabbit ball is a lively ball, one that scoots along; a gopher ball is a pitch that is hit for a home run. On the other hand, Ukrainian names of chess pieces, such as ‘ele- phant’ for the bishop (in English formerly called archer, and in still earlier times alfin from Arabic al-fil ‘elephant’) and ‘horse’ (for the knight), preserve imprints of a war scenario that originally underlay the game. Lexical variation in animal-related sports terminology includes lexical gaps in the other language, often because the sport is not practiced in Ukraine (baseball, golf) or in the English-speaking world (hopak), respec- tively. There are also non-correspondences in motivation of the terms denot- ing the same entities, e.g. a horizontal stand is the English correlate of the Ukrainian gymnastics term ‘swallow’ but the former does not ap- peal to animal imagery; or references to different animals, e.g. butterfly – ‘dolphin’. Overall, figurative animal names are not very common in modern profes- sional sports terminology due to the lack of transparency in their denotative meanings and apparently undesirable connotations. In the globalizing world, professional sports terminology is subject to formalization and stylistic neu- 169 tralization for successful international communication. However, there is always enough room for sports jargon, which proves to be culture specific. 5 Conclusion In this paper we have argued that conventional metaphors should not be dis- regarded as impotent, since they provide access to cognitive tendencies in conceptualization and ontologization of knowledge in the language structure and can be seen as the linguistic repository of a number of cultural practices that are common or specific to a given language. Both English and Ukrainian animal metaphors realized in lexemes, collo- cations and idioms are not isolated but form certain networks which can be interpreted via algorithmic descriptions of relations organized in scenarios. The latter are discretely connected to collective cultural knowledge of lan- guage communities and carry imprints of culture-specific practices from the present as well as the past. The animal component in modern sports-related terminology reflects the influence of the naïve perception of the world as regards both language structure and content. The contrastive analysis of conventional animal metaphors related to sports has shown that although conceptual mappings in the domains under study vary in English and Ukrainian and that differences also exist as regards highlighting various parts of similar scenarios, all in all, there is more diver- gence in the lexicalization of conventional animal metaphors in the lan- guages concerned than there is in their conceptualization, due to the common European cultural heritage of the English-speaking and Ukrainian-speaking communities. References ‘Animals in Sport’. On-line at: http://en.wikipedia.org/wiki/Animals_in_sport (re- trieved on 22 August 2008). Baranov, Anatolij N. & Dmitrij O. Dobrovol’skij. 1996. ‘Cognitive Modeling of Actual Meaning in the Field of Phraseology’, in Journal of Pragmatics 25: 409– 429. Clark, Herbert H. 1996. Using Language. Cambridge: Cambridge University Press. Croft, William. 2000. Explaining Language Change. An Evolutionary Approach. Harlow: Longman. Deignan Alice. 2005. Metaphor and Corpus Linguistics. Amsterdam: John Benja- mins. Fillmore, Charles J. 1985. ‘Frames and the Semantics of Understanding’, in Quad- erni di semantica 6: 222–254. Gibbs, Raymond W. Jr. & Gerard J. Steen (eds). 1999. Metaphor in Cognitive Lin- guistics: Selected Papers from the Fifth International Cognitive Linguistics Conference, Amsterdam, 1997. Amsterdam: John Benjamins. 170 Goatly, Andrew. 1997. The Language of Metaphors. London: Routledge. Griffin, Emma. 2007. Blood Sport: Hunting in Britain since 1066. New Haven: Yale University Press. Koenig, Jean-Pierre, Gail Mauner & Breton Bienvenue. 2008. ‘What with? The Anatomy of a (Proto)-Role’, in Journal of Semantics 25(2): 175–220. Lakoff, George. 1987. Women, Fire and Dangerous Things: What categories reveal about the mind. Chicago: University of Chicago Press. Lakoff, George & Mark Johnson. 1980. Metaphors We Live By. Chicago: University of Chicago Press. Langacker, Ronald W. 1987. Foundations of Cognitive Grammar: Theoretical pre- requisites. Stanford, CA: Stanford University Press. Lease, Gary. 2008. ‘Hunting and the Origins of Religion’, in Taylor, Bron R. (ed.), The Encyclopedia of Religion and Nature. London: Continuum, 805–809. Lewis, David. 1969. Convention. Cambridge, Mass: MIT Press. Ortega y Gasset, José. 1972. Meditations on Hunting (tr. Howard B. Wescott). New York: Scribner. Palmer, Gary B. 1996. Toward a Theory of Cultural Linguistics. Austin: University of Texas Press. Panther, Klaus-Uwe & Linda L. Thornburg. 2002. ‘The Roles of Metaphor and Metonymy in -er Nominals’, in Dirven, René & Ralf Pörings (eds), Metaphor and Metonymy in Comparison and Contrast. Berlin: Mouton de Gruyter, 279– 319. Petersen, David. 2008. ‘Hunting Spirituality and Animism’, in Taylor, Bron R. (ed.), The Encyclopedia of Religion and Nature. London: Continuum, 811–813. Pylat, Volodymyr. 1999. Combat Hopak. Lviv (Ukraine): Halyts’ka Vydavnycha Spilka. Rosch, Eleanor. 1975. ‘Cognitive Representation of Semantic Categories’, in Jour- nal of Experimental Psychology 104(3): 192–233. Schank, Roger & Robert Abelson. 1977. Scripts, Plans, Goals and Understanding: An inquiry into human knowledge structures. Hillsdale, NJ: Lawrence Erlbaum Associates. Svanlund, Jan. 2007. ‘Metaphor and Convention’, in Cognitive Linguistics 18(1): 47–89. Wierzbicka, Anna. 1994. ‘“Cultural Scripts”: A new approach to the study of cross- cultural communication’, in Pütz, Martin (ed.), Language Contact and Lan- guage Conflict. Amsterdam (Holland): John Benjamins, 69–87. Zhajvoronok, Vitalij. 2006. Znakв Ukrains’koi Etnokul’turв [Signs of Ukrainian Ethnoculture]. Kiev: Dovira. Data sources Ammer, Christine (ed.). 1997. The American Heritage Dictionary of Idioms. Boston: Houghton Mifflin. Barantsev, Kostiantyn T. 2005. Anhlo-Ukrains’kвi Fraгeolohichnвi Slovnвk [Eng- lish-Ukrainian Phraseological Dictionary]. Kiev: Znannia. Collins English Dictionary. 2006. 8th ed. Glasgow: HarperCollins. Evans, Ivor H. (ed.). 1981. Breаer’s Dictionary of Phrase and Fable. 14th ed. New York: Harper Perennial. Horets’kyi, Petro J. & Andrij A. Burjachok. 1970–1980. Slovnвk Ukrains’koi Movв [Dictionary of the Ukrainian Language]. 11 vols. Kiev: Naukova Dumka. 171 Lyman, Darryl. 1983. Animal Things We Say. New York: Jonathan David Publish- ers. Mel’nychuk, Oleksandr S. (ed.) 1983–2006. Etвmolohichnвj Slovnвk Ukrains’koi Movy [Etymological Ukrainian Dictionary]. 7 vols. Kiev: Naukova Dumka. Oxford English Dictionary. 2004. Second Edition on CD-ROM. Version 3.1. Ox- ford: Oxford University Press. Palmatier, Robert A. 1995. Speaking of Animals: A Dictionary of Animal Meta- phors. Westport, Conn.: Greenwood Press. Palmatier, Robert A. & Harold L. Ray. 1989. Sports Talk: A Dictionary of Sports Metaphors. Westport, Conn.: Greenwood Press. Vynnyk, Vasyl’ O. 2003. Slovnвk Phraseolohismiv Ukrains’koi Movв [Dictionary of Phraseologisms of the Ukrainian Language]. Kiev: Naukova Dumka. 172 Ana Roldán-Riejos Therapeutic Metaphors in Engineering: How to Cure a Building Structure Abstract. Cognitive linguists have taken pains to point out the pervasiveness of conceptual mappings, particularly conceptual blending and integration, that underlie language and that are unconsciously used in everyday speech (Fauconnier 1997, Fauconnier & Turner 2002; Rohrer 2007; Grady, Oakley & Coulson 1999). More- over, as a further development of this work, there is a growing interest in research devoted to the conceptual mappings that comprise specialized technical disciplines. Lakoff & Núñez 2000, for example, have produced a major breakthrough on the understanding of concepts in mathematics, through conceptual metaphor and as a result not of purely abstract concepts but rather of embodiment. On the engineering and architectural front, analyses on the use of metaphor, blending and categorization in English and Spanish have likewise appeared in recent times (Úbeda 2001, Roldán 1999, Caballero 2003a, 2003b, Roldán & Ubeda 2006, Roldán & Protasenia 2007). The present paper seeks to show a number of significant conceptual mappings un- derlying the language of architecture and civil engineering that seem to shape the way engineers and architects communicate. In order to work with a significant seg- ment of linguistic expressions in this field, a corpus taken from a widely used tech- nical Spanish engineering journal article was collected and analysed. The examina- tion of the data obtained indicates that many tokens make a direct reference to thera- peutic conceptual mappings, highlighting medical domains such as “diagnosing”, “treating” and “curing”. Hence, the paper illustrates how this notion is instantiated by the corresponding bodily conceptual integration. In addition, we wish to under- line the function of visual metaphors in the world of modern architecture by evoking parts of human or animal anatomy, and how this is visibly noticeable in contempo- rary buildings and public works structures. Keywords: ESP cognitive approach; metaphor; blending; conceptual integration; engineering and architectural representations 1 Introduction When I started working in the academic world of civil engineering some years ago and I began to become familiar with their jargon in both English and Spanish, I noticed some striking features. For example, a good number of linguistic terms were borrowed from the medical domain. They were not just related, they were the same. My first impression soon became a fact 173 when I started gathering written (books, journals, manuals) and spoken ma- terial (interviews and lectures by engineers) about this subject matter. One of the best-selling books in the university where I work has the title of Patología de las Estructuras ‘Pathology of Structures,’ by Prof. Calavera, a respected Spanish civil engineering scholar. As a matter of fact, courses labelled with this or similar names from the field of medicine are not at all unusual. Websites named “Building or Construction Pathology” can be easi- ly found on the Internet. Likewise, I learnt that engineering activities involve the use of auscultation devices for dams, that bleeding is an undesirable ef- fect in different types of concrete and that metal beams may suffer from stress and fatigue. All this confirmed that some metaphoric mappings are common in engineering. For engineers, the use of these conceptual mappings has become completely entrenched in their way of thinking, reasoning and communicating, and most of the time they are not even aware of it. Therefore, after this verification, the next step was to learn the reasons why some specific mappings and metaphors were more salient in both Eng- lish and Spanish civil engineering. This task seemed to be worthwhile in order to acquire a better understanding of engineers’ ways of thinking and categorizing. After all, as Fauconnier remarks: “Language is only the tip of a spectacular cognitive iceberg” (1999). Typically, engineers’ jobs include the design of big structures, as well as solving problems that may affect these structures. For example, they are concerned with how to solve the problem of crossing a river by the construc- tion of the most suitable bridge or how to link two distant towns by means of a highway. Bridge constructions, for instance, must meet standard criteria and fulfil technical conditions such as the ability to withstand opposing forc- es and bear various types of loads over their decks. On the other hand, any bridge is situated at a particular place and surrounded by a unique environ- ment. Building a road bridge is not the same as building one across a bay. One solution to help manage the complex and sophisticated techniques in- volved in bridge construction is to treat bridges as living creatures, i.e. hav- ing a lifespan and a type of behaviour/performance, and hence to monitor their physical condition through convenient technical methods. Accordingly, bridges’ health should be regularly checked to avoid major mishaps due to fatigue, decay or stress. This type of reasoning is done in a conventional way, similar to the way that we use and understand idioms. For instance, if someone advises us about not “throwing in the towel”, we do not actually think of a towel, or about being in a boxing ring. Instead, we understand that we are being en- couraged to overcome some sort of hardship. In the same way, during their work, engineers are not consciously thinking of the borrowed mappings they are using. They just apply them automatically, since this technique has been previously assimilated during their training. The moment they are exposed to this type of language, everything clicks into place for them. 174 It is true, however, that medical practice and engineering share certain characteristics, including a similar pragmatic approach to the job. In today’s world, engineers have to deal with uncertainties and risks; they have to apply probabilistic theories and consider a lot of variables when making decisions. Both engineers and physicians know that they are dependent on obtaining reliable contextual and perceptual information and must rely on technical tools to get data (Blockley 2005: vii–viii). In Calatrava’s words, engineers are concerned with the “empiric, the experimental understanding of the reali- ty” (2008: BBC interview). As in medicine, civil engineers frequently hold the lives of others in their hands; therefore the engineering profession in- cludes learning from errors. For example, the Tacoma Narrows bridge col- lapse in the state of Washington remains a prototypical case study for engi- neers of why suspension bridges may fail and result in loss of life. Therefore, examples like this are exhaustively studied for future prevention by means of laboratory or field analyses or by performing autopsies of defunct construc- tions (Pathology Construction Website). The main aim of the present paper is to focus on the use of medical lan- guage as a major input for expressing civil engineering concepts. It is true that there are cases of civil engineering terms actually used as source do- mains to convey abstract concepts, as in cementing a friendship, or in collo- quial English hitting/going through the roof (becoming furious); or in the glass ceiling (applied to women’s career obstacles). Some uses may actually target the medical domain, as in colloquial Spanish: Estoy para el desguace, literally: ‘I am ready for the scrapyard,’ but actually meaning: ‘I’m feeling shattered or in terrible condition’. We will not consider this type of examples in this study, however; instead we will concentrate on inputs from the thera- peutic domain onto the engineering one. The conceptual integration framework proposed by Fauconnier (1997) and developed in Fauconnier & Turner (2002) is followed, as the most ap- propriate framework for our purposes, because it provides a more complete model than earlier metaphor theories. It is considered more unifying, be- cause the conceptual integration stance encompasses conceptual and image metaphor, blends, categorizations, frames, counterfactuals and metonymies. Examples of medical blends in engineering will be shown, as well as exam- ples to illustrate the importance of perception (visual representations) in engineering, including “image blends” in various descriptive examples, as analysed below. 2 Initial corpus collection A preliminary phase of this study consisted of collecting a corpus of engi- neering keywords and their main collocations, to create a representative sample (Roldán & Protasenia 2007). The preliminary aim was to identify 175 Figure 1. Frequencies of the term control. Figure 1. Frequencies of the term control. engineering words related to the medical domain. The corpus is comprised of 81 journal articles of Revista de Obras Públicas, which is the authorized professional journal for chartered Spanish civil engineers (Ingenieros de Canales, Caminos y Puertos). The articles are from January 2000 to Decem- ber 2004. Concordances, frequencies, clusters and keywords were subse- quently extracted and analysed by means of AntConc 2006 software and by applying the OU CREET procedure for metaphor analysis in discourse. Oth- er corpus approaches for identifying metaphor such as those carried out by Charteris-Black (2004), Caballero (2003a, 2003b), and Deignan (2005) also served as references when undertaking this work. Indeed, the widespread use of metaphor in architecture or in civil engi- neering has been already noted in Spanish and English (Caballero 2003a, 2003b, Úbeda 2001, Roldán & Ubeda 2006). Here, as in many other realms of language, metaphor appears to be pervasive. The civil engineering corpus collected supports this idea, since a considerable number of terms in the sample have links with human body input. Interestingly, the highest frequen- cy of terms, as shown in Figure 1 above, corresponded to control, and a closer examination of the concordances of this term suggests that the mean- ing of control is very similar to monitoring, i.e. when the engineer checks a structure to follow up its behaviour (Figure 2, next page). 176 Figure 2. Concordance of the term control. The equivalent mapping in the medical domain is checking someone to prevent or to cure a possible illness. This parallelism emphasizes the im- portance of embodiment mappings in civil engineering, and as we will see below, embodiment will also be characteristic in other conceptual operations such as counterfactuals, metonymy, frames and categorizations. Therefore, as stated above, we maintain that given the complexity and the variety of conceptual and perceptual mappings that can be detected in engineering communication, the science of blending is the best tool currently available. Hence mental space theory and blending (conceptual integration) theory is particularly well suited for explaining the construction of meaning in engi- neering. In blending or conceptual integration, selective projections are made from the target and the source inputs. The links between the inputs and the result- ing blend usually create an emergent innovative structure that can be inter- preted through patterns of compression or decompression (Fauconnier & Turner, 2008). The inputs consist of mental spaces that are conceptual con- tainers of basic information. They are connected to frames, which are sche- matic structures based on experience and related to memory and to previous knowledge. Basically, we can find at least four connected mental spaces: two inputs linked by cross-space mappings, a generic space, and the resulting 177 blend that contains an emergent structure different from the generic space and the inputs. A series of operations may be involved, such as composition, i.e. a selection from the inputs to the blended space; completion, i.e. com- pleting the scenario with previous or background information; and elabora- tion, i.e. creating (emerging) structure not predictable from the inputs. The cross-space mappings connect matches and their counterparts in the inputs. One paramount aspect about blending is that it is a dynamic process, since novel mappings are constantly emerging; it is like some sort of work in con- struction, unfinished and ongoing. And it is through cognitive patterns such as compression and decompression that information is processed and put to use. As human beings, we possess frames, background knowledge, long- or short-term memories of scenarios, situations and experience that can be acti- vated and updated at any single point when necessary. In engineering, there are specific conventional mental spaces that can be activated. These mental spaces consist of particular frames which when connected with other spaces may result in integration networks that will deliver emergent structures. For example, the “engineer respecting the environment” frame and the “engineer taking care of people” frame, which will be seen below, are quite recent and innovative in engineering. The emergent structure of this particular blend is currently very productive. From the complex interrelations of blending net- works and vital relations, compressions and decompressions operate in the emergent structure. The networks may show links between the inputs, which according to Fauconnier & Turner (2002: 92–93) correspond to “outer-space links”, with those inside the blend accounting for “inner-space links”. For example, in Figure 3 the mappings show the cross-space mapping links from the borrowed medical roles and frames of physicians, diseases, symptoms, diagnoses, cures, and treatments applied to the engineering roles and frames that comprise bridges affected by structural problems, in this case fatigue. There is a generic space that is shared, i.e. the existence of a problem and the need to solve it by following a pragmatic and empirical approach shared in the medical and the engineering mental space inputs. However, since they are two very different scenarios (i.e. a surgery or a hos- pital vs. the specific location of the bridge), contrasting roles and dissimilar symptoms, emergent structure is created in the resulting blend. The medical domain acts as a sort of scaffolding to prop up ad hoc structures required in the engineering domain. The selected repertoire is not whimsical, but pre- cise. It fits into a specific pattern, covering only various engineering aspects, like fatigue or stress as diagnostic symptoms and rehabilitation as the sug- gested treatment. This means that the input that is coherent at the human scale (the medical input) provides coherence to the other input (engineering), so that the former has been compressed when projected into the blend. The symptoms (fatigue), which are also in the blend, have to be decompressed by the doctor-engineer, since the cause of the illness-problem is compressed in 178 Problem to be Generic space solved: similar Bridge Patient pragmatic affected affected approach and by by dis- observation fatigue ease methods Elements Elements Patient, Bridge, Doctor, Engineer, Symptoms Doctor Bad per- Engineer formance Relations Cure Relations (Person, Blend Bridge Repair, cure Patient, rehabili- (Person, Surgery) tated and Bridge, cured Environ- ment) Figure 3. Bridge rehabilitation the specific symptoms. Inferences result because of the interacting links in the inner spaces of the blend, delivering possible solutions for the bridge, such as its rehabilitation. This includes the environment where the bridge is situated. So, in the final blend we find the following conceptual relations: cause-effect (because of the intervening action), change (from initial state A to final outcome B), identity (roles of doctor-engineer) as well as part/whole relation links (only some elements are highlighted: e.g. the fatigue problem). These elements are compressed either by scaling (the condition of the struc- ture can be better or worse) or by syncopation (only some aspects of the process affecting the structure are relevant enough to be activated in the blend, e.g. its lack of strength). In addition, there may be two alternatives in the condition of the bridge with structural problems. One is the current faulty condition and the other the desired counterfactual state which the engineer wishes the bridge to achieve when repaired. For example, if it is experiencing a settling problem because of poor subsoil, then the foundations must be reinforced. In one blend the structure is unstable and weak. In the second case, the engineer has solved the instability problems. Most of the time, engineers work with prob- abilistic methods and calculations, not deterministic ones, and therefore a lot of uncertainty and risk exist in engineering decisions. There is not just one right or correct answer to the problems that engineers face, so they are 179 trained to try different possibilities to resolve difficult or doubtful issues. In the blend that was analysed above, the consideration of playing the medical role to attend a patient (the structure) is only evoked, not explicitly felt, be- cause the blend forms part of the engineer’s expert training, acquired during the whole training process involved in becoming an engineer. 3 Engineers’ blends and networks In Fauconnier & Turner (2002: 129), various types of networks are devel- oped. For example, there are single-scope networks, where interconnected links occur in two domains. The example they proposed is cross-mapping domains from boxing input and economics input, which convey inferences between fighting-struggle and business. As Fauconnier & Turner put it, de- spite the strong asymmetry between the inputs, the resulting network gives us the feeling that one of them (boxing) provides insight into the other (eco- nomics). If we translate this point to our discussion, the inferences are that an engineering structure needs medical attention, because of embodied prob- lems/symptoms such as: - bleeding - stress/strain - fatigue - collapse The resulting blend has inherited structure from the medical input and im- plicitly includes the conventional metaphor of structures as human beings, which eases the task of establishing links between both inputs. However, the two main agents in this network, the doctors and the engineers, are not actu- ally counterparts in the frame of the inputs: the engineers are not exact coun- terparts of the doctors, but rather of what they do, i.e. solving problems. In the one case they look after big structures and in the other they take care of patients. As shown in Figures 4 and 5, there are multiple spaces and complex conceptual integration networks in engineering. An engineer has in mind many spaces and frames; some of them may limit the scope of their projects and act as constraints in the construction process. Some of these spaces are: - that of the materials to use or previously used - that of aesthetics - that of building, local and urban regulations - that of mathematical calculations - that of geological conditions - that of environmental laws 180 Structures Medical People in bodies People in structures Figure 4. Interrelated medical-engineering networks - that of the budget - that of security for people - that of the team of people to work with - that of the culture and language where the structure will be located. At all times they must consider the effect of their work on the environment and also the whole life impact of the structure on the outside world. For ex- ample, in the mental spaces employed when designing a bridge, the main conceptual relations are time, cause–effect, intentionality, change, identity, and representation. The time relation exists in the sense that building or re- pairing a structure takes time. There is also cause and effect, from the initial state A to the final product B, as well as the intentionality to do so. This also includes a qualitative and quantitative change from the first condition to the end-result condition, having a series of successive stages in between. Finally, there is what the structure represents (representation). Therefore, when con- ceiving of a structure, there is an anticipation of its good or bad potential performance. Additionally, one major factor that the engineer must consider is the effect that the structure will have on people. As seen above in the en- gineer-as-a-doctor blend, the emergent structure compressed in the blend entails the metaphor of the structure as being human, i.e. the “humanization” of the structure, and therefore deserving of care. But going one step further, there are links connecting the welfare of people in bodies—which is sought after (responsible: doctors)—and the welfare of the people in the structure (responsible: engineers), as shown in Figure 4. From this emergent structure, the degree of satisfaction of the people using the structure is inherited, a 181 Exterior Internal appearance well-being Enjoyment Aesthetics of people Figure 5. Further networks on the medical-engineering blend welfare which is not only physical, but also aesthetic (Figure 5). This is a double-scope network (Fauconnier & Turner 2002: 132) because the differ- ent organizing frames of the inputs contribute to the blend in an innovative way. For example, the welfare of people in a building is about people feeling comfortable in it, including the right quantity of light (not too bright or too dim) and a pleasing visual environment through the design of windows and lighting systems; the absence of distracting noise; the right temperature and ventilation. Likewise, when the external appearance of the building is aesthetically pleasant, this may facilitate the internal well-being of people in the building (better health) as well as their enjoyment. Actually, if this aim is achieved in different structures, and care is taken as regards the earth’s resources and pollution control, this would have the desired effect of living on a healthy planet. In Figure 5 we see the links connecting the networks between the external look of the structure and the internal well-being of the people who use the structure and between the aesthetic appearance of the structure and the enjoyment of people. 182 4 Images and perceptions in engineering As Blockley (2005: vii) notes, the different stages of the design of an engi- neering structure cannot be disconnected and they are mainly envisaged in the form of images. It appears that engineers visualise their projects in imag- es before putting them into words. The professional builds up a visual repre- sentation of the entire process right from the beginning until its termination. For example, Figure 6 shows the prototypical visual representation of a cable-stayed bridge (a) and of a suspension bridge (b), as well as the distri- bution of forces. To explain technicalities and structural principles concern- ing these bridges, engineers prefer to use images like these rather than using words or words alone. A Spanish engineer could communicate fairly well with a German engineer in this way without speaking each other’s lan- guages. Figure 6 (a). Cable-stayed bridge: prototypical representations Figure 6 (b). Suspension bridge: prototypical representations 183 As shown in previous research (Caballero 2003a, 2003b; Roldán & Úbeda 2006), there are a considerable number of terms in architecture and engineer- ing construction that evoke a visual origin and description. To portray their designs and creations, engineers and architects prefer to use drawings and figures that are mostly images and that therefore activate visual interpreta- tion in the beholder. Hence, when wishing to describe their work in words, they choose pictographic phrases, like the jagged fan of five overscaled con- crete fins (Caballero 2003b: 150). Despite rivalries over professional compe- tences and responsibilities, if there is one thing that architects and engineers have in common, it is the visual thinking of their work. In interviews Santia- go Calatrava, the well-known Spanish architect and civil engineer, has acknowledged about his own modus operandi: You see indeed I, I, I draw very much, constantly, among other things because in the communication you see a drawing or you see or a sign is more im- portant than one thousand words so. ... Drawing is the laboratory of my ideas, their first expression. My hand motions speak even before my mouth is aware of them. (Calatrava s. d.; Bet Levi 2008) In a way, an important (and difficult) skill that both architects and engineers must acquire during their training is verbalising, putting their mental images into spoken and written words, which probably explains why metaphor or metonymy is so common. Caballero (2003a, 2003b) has examined the occur- rence of conceptual metaphor and, particularly of image metaphor, in archi- tecture. Many of the examples that she presents from her collected corpus have a visual origin and nature. In addition, she points out that it is often not so easy to differentiate conceptual metaphor from image metaphor in archi- tecture texts, claiming that there is often interplay between the two types, because of “the visual and aesthetic constraints of the discipline” (2003b: 150). Adding to that, we would also like to underline the key role of the vis- ual component in engineering, as well as the (frequently underestimated) aesthetic element. In fact, this issue can be tested in many historic examples, such as the Segovia Aqueduct or the Great Wall of China. In Roldán & Úbe- da (2006: 538), additional evidence indicates a considerable use of meto- nymic images in descriptive engineering construction texts. Another feature in common for architects and engineers is the frequent presence of embodiment in their conceptual mappings and, as we will show in the examples below, in image blending. This is also reflected in their use of metaphors from the medical domain. Quoting Calatrava again: Yes, and it means you see that in your body almost is architecture, so indeed you see if you put your hands together you see or, or your face you see the expression of your mouth, you see even your, the proportions of your body, all those things you see has an archi..., let’s say are patterns of understanding of architecture. [s.d.] 184 Figure 7. The “Turning Torso” building (Malmö) and its analogy with the hu- man body This way of reasoning explains the ubiquity of the human figure in engineer- ing metaphorical mappings and blends, part of which we explained above in the STRUCTURES AS PATIENTS and ENGINEERS AS DOCTORS mappings. This fea- ture arises from considering structures as being as fragile as human beings and therefore requiring medical attention. Let us look at how these blends operate together in images. We can see in Figure 7 an example of how the moving shape of the human body inspired Calatrava in the “Turning Torso” high-rise building in Malmö (Sweden). From the input of the images of human body movements stored in his memory, and from the input of his knowledge as an architect and engineer, Calatrava has been able to design this building. Here he has succeeded in fusing in the building image the conceptual blends of the “anthropology” of the building (its aesthetic aspect) and the engineering structural principles that make it inhabitable. In the earliest stages of a building’s design the en- gineer will use sketches, computer models and calculations to seek out the best solution for the particular building. Such aspects as the direction that the building faces, the materials that are used to construct it, the types of glass and the interaction with the local climate will all need to be modelled and optimized and this is done with images. In the example in Figure 8 (next page), we can see an eye-catching image demonstrating with actual people the load distribution principles applied in the Forth Rail Bridge of Scotland (built in 1887). The sketch above the men shows the weight of the central span of the bridge being transmitted to the banks through diamond-shaped supports. The central “weight” is Kaichi Watanabe, a Japanese engineer; Sir John Fowler and Sir Benjamin Baker, the bridge designers, provide the sup- ports. 185 Figure 8. Load distributions in the Firth of Forth Bridge (Scotland, UK) 5 Image blending: representation and interpretation In addition, the bridge designer will sometimes attempt to represent a super- imposed image in the actual shape of the bridge. In these cases, it is not only the utility of the bridge that matters, but also its iconic representation. Calatrava, for instance, frequently elaborates his works as artefacts and sculptures. Thus the Alamillo Bridge (Figure 9), designed as a portico to the 1992 Seville Expo, has been commonly perceived and interpreted as a harp, a fan and a swan over the Guadalquivir River. Therefore, apart from its pro- totypical use as a bridge to cross a river, the iconic interpretation of the bridge depends upon the conceptual relations of representation, analogy and intentionality. There are a whole set of conceptual relations, however, that go back to the moment when there was no bridge, only blocks of concrete to be construct- ed. We can see an evolution, i.e. how some blocks of concrete can become a unique and practical structure. Let us consider the relations involved in the process:   Change: transformation from one state to the final product. Cause-effect: The size changes, the structure is created as such: there  is a compression implied. Identity: it is the same concrete, but now with a different shape (com- pressed with time produces uniqueness in the end). 186 Figure 9. The Alamillo Bridge, Seville   Time: connected to the cause-effect and change relations. Part-whole: the concrete blocks are transformed into a bridge (cause-  effect). In the blend, they are fused.  Space: the concrete was transported to its final place over the river. Role: related to its value—the role of the bridge was to be the gate-  way to the 1992 Seville Expo, i.e. a symbol of the city. Representation (the thing represented): the Alamillo Bridge, not the  Golden Gate Bridge. Analogy/Similarity: A harp and a fan—this has the compression of  the aesthetic effect (cause-effect and role-value). Category: started as analogy and becomes compressed into a category  relation: it is a fan, the Malmo building is a torso.  Property: A safe bridge—it causes you to feel safe when crossing it. Intentionality: The designer desired and sought the effect of creating a city symbol. From all the activated conceptual relations, let us consider the relations of analogy/similarity, role-value and cause-effect occurring in this example. From the analogy projected from the inputs of the shape of the bridge like a swan, there is the resulting compression of the bridge as an aquatic bird and the aesthetic blend of the beauty of the swan, which metonymically becomes the beauty of the bridge. The mapping in the blend is iconic because the 187 Figure 10. The Katehaki Bridge, Athens form (the swan) reflects the meaning (the bridge). It also reflects the inten- tionality of its design to turn the bridge shape into a symbol. In the case of the representation of the bridge as a harp, there are additional other com- pressed effects, such as the musical sound of the water. The interpretation of the bridge as a fan entails folkloric local elements associated with the host city, Seville, which are added to the compression. Calatrava himself has acknowledged that he envisages the bridge as a dialogue between the deck and the pylon sustaining it. Let us consider now another bridge, located in Athens (Greece), the Katehaki bridge, conceived by Calatrava as an ancient Greek vessel, as shown in Figure 10 above. The image blend of the bridge as a vessel that transports people over the river is achieved, since it is only a pedestrian bridge. On the other hand, this is an ancient Greek vessel built in Athens, which is the metonymic compres- sion that Calatrava intended to convey, since [i]f the bridge is really successful, we can identify the place, and sometimes even the city itself, by means of it, as in the case of the Golden Gate Bridge in San Francisco, for example. (Bet Levi 2008) 188 Finally, as we have seen above, engineers operate with multiple spaces and blends, among which the medical domain is an important productive one. These conceptual integrations can be automatically activated when needed. The engineer must display what Fauconnier & Turner call global insights (2002: 84). This phenomenon also occurs in sports. For instance, a basket- ball player knows about different strategies for shooting or defence “by heart”, because of previous and intensive training. So the player must have a good knowledge of cause-effect consequences of the different moves exe- cuted when shooting or just passing the ball. The results will depend upon the level of expertise of the players. The engineer must be prepared to antic- ipate different situations too. This leads to “expert performance”, and to the application of global insights, where the cause-effect progression should be crystal clear. 6 Conclusions This paper has explored the significance and the scope of medical metaphor- ic mappings and conceptual blends in civil engineering practice. Similarities, analogies, disanalogies and other vital conceptual relations are revealed by a set of correspondences between entities in the two domains. Despite their contrasting frames, elements and roles, both engineers and doctors have to allow for errors in their practice. Engineering is not an exact science and neither is medicine. Both professionals are practitioners trying to calculate unpredicted factors. Likewise, an important feature of engineering practice is making an ex- tensive use of images (pictures, sketches, diagrams, graphs) when thinking and reasoning, hence the widespread occurrence of image-metaphors in their conversations and texts. The representation—and interpretation—of images appears to be more direct, immediate, holistic and eye-catching than the use of words, hence the use of iconic designs and creations. It is suggested here that within the field of civil engineering, not only are images a complement of words, but also words’ main function seems to be the description of imag- es. 189 References Bet Levi, Roi. 2008. ‘Deconstructing Calatrava’, in Haaretz Daily Newspaper. On- line at http://www.haaretz.com/deconstructing-calatrava-1.246381 (retrieved on 17 August 2008). Blockey, David. 2005. New Dictionary of Civil Engineering. London: Penguin Books. Caballero, Rosario. 2003a. ‘Talking about Space: Image metaphor in architectural discourse’, in Annual Review of Cognitive Linguistics 1: 87–105. —. 2003b. ‘Metaphor and Genre: The presence and role of metaphor in the building review’, in Applied Linguistics 24/2: 145–167. Calatrava, Santiago. s.d. ‘Interview Transcript on BBC-Radio 3’. On-line at: http://www.bbc.co.uk/radio3/johntusainterview/calatrava_transcript.shtml (re- trieved on 17 August 2008). ‘Construction Pathology’. s.d. On-line at: http://www.angelfire.com/biz/Building Pathology/index.html (retrieved on 23 January 2005). Charteris-Black, Jonathan. 2004. Corpus Approaches to Critical Metaphor Analysis. Basingstoke: Palgrave Macmillan. Deignan, Alice. 2005. Metaphor and Corpus Linguistics. Amsterdam: John Benja- mins. Fauconnier, Gilles. 1999. ‘Methods and Generalizations’, in Janssen, Theo & Gisela Redeker (eds), Cognitive Linguistics: Foundations, scope and methodology. Berlin: Mouton De Gruyter. Also on-line as ‘Introduction to Methods and Gen- eralizations’ at: http://cogweb.ucla.edu/Abstracts/Fauconnier_99.html (retrieved on 9 April 2005). —. 1997. Mappings in Thought and Language. Cambridge: Cambridge University Press. Fauconnier, Gilles & Mark Turner. 2008. ‘Rethinking Metaphor’, in Gibbs, Ray- mond W. Jr. (ed.), The Cambridge Handbook of Metaphor and Thought. Cam- bridge: Cambridge University Press, 53–66. —. 2002. The Way We Think: Conceptual blending and the mind’s hidden complexi- ties. New York: Basic Books. Grady, Joseph, Todd Oakley & Seana Coulson. 1999. ‘Blending and Metaphor’, in Steen, Gerard J. & Raymond W. Gibbs, Jr. (eds), Metaphor in Cognitive Lin- guistics. Amsterdam: John Benjamins, 101–124. Lakoff, George & Rafael E Núñez. 2000. Where Mathematics Comes From: How the embodied mind brings mathematics into being. New York: Basic Books. Patjawit, Anun & Worsak Kanok-Nukulchai. 2005. ‘Health Monitoring of Highway Bridges Based on a Global Flexibility Index’, in Engineering Structures 27: 1385–1391. ‘Procedure for Metaphor Analysis: Analysing metaphors in discourse structure’. Open University, UK. On-line at: http://creet.open.ac.uk/projects/metaphor- analysis/procedure.cfm (retrieved on 3 July 2008). Rohrer, Tim. 2007. ‘The Body in Space: Dimensions of embodiment’, in Ziemke, Tom, Jordan Zlatev & Roslyn M. Frank (eds), Body, Language and Mind. Ber- lin: Mouton de Gruyter, 339–378. Roldán-Riejos, Ana. 1999. ‘Applications of Cognitive Theory to Interdisciplinary Work in Languages for Specific Purposes’, in Ibérica 1: 29–37. Roldán-Riejos, Ana & Paloma Úbeda-Mansilla. 2006. ‘Metaphor Use in a Specific Genre of Engineering Discourse’, in European Journal of Engineering Educa- tion 31 (5): 531–541. 190 Roldán-Riejos, Ana & Yulia Protasenia. 2007. ‘Figurative Language in Academic and Professional Engineering’, in VI Congresso da AELFE. CD Proceedings (University of Lisbon, 13–16 September 2007). Úbeda Mansilla, Paloma. 2001. Estudio de un corpus de textos conversacionales en inglés realizado en estudios de arquitectura: su aplicación al diseño de un pro- grama de inglés para arquitectos [A Study of a Corpus of English Conversation among Students of Architecture: Its application to the design of a study program in English for architects]. Madrid: Publicaciones de la Universidad Com- plutense. Illustration acknowledgements Figure 6 a: http://en.wikipedia.org/wiki/Cable-stayed_bridge b: http://www.pbs.org/wgbh/buildingbig/bridge/susp_forces.html Figure 7 a: http://www.casamexico.com.mx/cm/noticia_detalle.php?id=1897 b: http://blog.miragestudio7.com/wp- content/uploads/2007/07/santiago_calatrava_turning_torso.jpg Figure 8 http://www.en.wikipedia.org/wiki/Forth_Railway_Bridge Figure 9 http://en.wikipedia.org/wiki/File:Calatrava_Puente_del_Alamillo_Seville.jpg Figure 10 http://farm4.static.flickr.com/3083/2568224425_90fc1ee9f1.jpg?v=0 191 Rotimi Taiwo Metaphors in Nigerian Political Discourse Abstract: Metaphors play a central role in public discourse. This is particularly true of political discourse, which is mostly persuasive and rhetorical in nature. Metaphors help to shape the structure of political categorisation and argumentation. This paper examines the use of metaphors in selected Nigerian political discourses. The analy- sis of metaphor in this paper was guided by Lakoff & Johnson’s (1980) theory of conceptual metaphors. Three target domains were identified as sources of conceptual metaphors in the data—the nation, politicians and politics. The NATION was concep- tualised as a FAMILY and a PERSON. There was also the conceptual mapping of the POLITICIAN AS BUILDER. We also identified the conceptual mappings of POLITICS AS BATTLE and POLITICS AS JOURNEY. The mapping principle of targets and domains identified in the data are used by Nigerian politicians to fulfill their persuasive and rhetorical goals in political discourse. Key words: metaphor, politics, discourse, rhetoric, Nigeria, conceptual 1 Introduction Language is a tool for expressing man’s social reality. Politics is one of the spheres of social life in which language plays a pivotal role. Politicians en- gage in different rhetorical uses of language to achieve their political goals. One of the obvious rhetorical strategies used in political discourse is indirect language. The use of indirect language is made possible by the changeable nature of word meanings, which Warren (1992: 128) describes as negotiable and dynamic. This nature of word meaning makes it possible to assign refer- ences that are not usually found in dictionary descriptions. Political language is associated with different rhetorical and figurative strategies like circumlo- cution, irony, symbolisms, innuendos, euphemisms, and metaphors (Bosman, 1987; Howe, 1988; Yusuf, 2003; Linfoot-Ham, 2005; Charteris- Black, 2005; Carver & Pikalo, 2008). Indirect language is often used when politicians have to talk about politically risky topics, and often directly linked to certain factors such as the protection of their careers, their desire to gain political and interactional advantage over their opponents, politeness, and so forth (Obeng, 1997). Vestermark (2007: 1) links the use of metaphors to persuasion and propaganda in politics. This present study examines the use of metaphor as a major feature in Nigerian political discourse. 193 2 Political Discourse The idea of political discourse is approached by different disciplines from different perspectives. Our concern in this paper is the linguistic perspective of political discourse. Defining political discourse will hinge on what one considers as politics. Politics has both wide and narrow senses. The simplest definition will be the one limited to the activities of institutions, such as po- litical parties, government and parliament, in the fulfilment of political obli- gations. Politics is also conceived of as a struggle to gain and retain power among members of these institutions (Beard, 2000: 36). According to Bayley (2008), political discourse is a wide and diverse set of discourses, or genres, or registers, such as: policy papers, ministerial speeches, government press releases or press conferences, parliamentary discourse, party manifestos (or platforms), electoral speeches, and so forth. They are all characterised by the fact that they are spoken or written by (or for) primary political actors— members of the government or the opposition, members of parliament, lead- ers of political parties, candidates for office. A political discourse therefore is discourse in any political forum, such as campaigns, parliamentary de- bates, interviews, speeches, writing and so forth. Chilton & Schaffner (1997: 212) identify a political discourse as any dis- course whose linguistic or other actions involve power or its inverse, re- sistance. Wilson (2001: 398) describes political discourse as language used in formal and informal political contexts with political actors, such as politi- cians, political institutions, government, political media and political sup- porters operating in political environments with political goals (cf. Moreno, 2008: 34). Despite the fact that linguists are primarily concerned about the way lan- guage is structured in political discourse, their approaches to political dis- course differ. Scholars in linguistic stylistics of the British Contextualism school are concerned with the style, systems, structures and functions of the language, positing that the relationship between a language and the social functions that its serves is reflected in the internal organisation of the lan- guage (Halliday & Fawcett, 1987: 199). Within this school, there is another set of scholars associated with the British Critical Linguistic School. Critical linguists aim at identifying the social meanings that are expressed through lexis and syntax and considering the role that language plays in creating and reinforcing ideologies (Fowler et al. 1979) According to Threadgold (2003: 13), this group was strongly influenced by the work of the linguist Michael Halliday and his systemic functional grammar. Although they also used sty- listic approaches borrowed from Chomskyan transformational linguistics, Fowler in particular added to his approach the work of Barthes and early French semiotics. Critical Discourse Analysis (CDA), an interdisciplinary field of study, later developed from the Critical Linguistic tradition. CDA 194 sees language as a form of social practice, combining the methods of diverse disciplines such as sociology, social psychology, ethnography, semiotics, cultural studies and linguistics, drawing on the contributions of scholars whose primary concern has not been language, such as Habermas, Foucault and Bourdieu. This present study will blend the methods of CDA with that of the Cogni- tive Linguistics approach originated by Lakoff & Johnson (1980) to discuss the metaphorical expressions found in the data. Cognitive Linguistics (CL) sees learning, creation and usage in language as being best explained in ref- erence to human cognition. In Cognitive Linguistics, metaphors are not just linguistic phenomena of ‘figures of speech’, but a matter of thought (Lakoff, 1986: 218). As Cienki (2005: 1) notes, research in CL has shown how meta- phorical expressions in language about one domain can reflect deep-seated thinking in terms of another. According to scholars in CL, much of our eve- ryday talk, as well as much of our thought and much of our reality, is struc- tured metaphorically. This means that most of our abstract categories are organised cognitively by structures borrowed from more concrete categories (Bednarek, 2005: 7). For instance, the conceptual metaphor TIME IS MONEY is reflected in the linguistic expressions you are wasting my time, this gadget will save you hours, is that worth your while?, he is living on borrowed time, etc. (Lakoff & Johnson 1980: 7–8). 3 Nigerian Political Discourse The Nigerian political discourse has largely been shaped by the stock of people holding political power at any point in time. Like many African states, Nigeria has been in and out of one form of political crisis or the other. The country became a sovereign state on October 1, 1960 and six years after, the military seized power from the civilian rulers and held it for thirteen years, during which time the nation witnessed a civil war with bloody coups and counter-coups. The military handed rule over to a civilian government in 1979, but this government lasted only four years before another highjack of power by the military. They ruled for another sixteen years before handing over power to the government of Olusegun Obasanjo in 1999. Since then, the civilians have been in power. Several issues shape discourse in the Nigerian political scene. Since her independence, Nigeria has been going through different kinds of socio- political experiences, most of which can be summarised under the following: corruption and mismanagement of resources, human rights abuses, ethno- religious violence, resource-related crises, highly flawed electoral processes, power generation crises, labour-related crises, insecurity of lives and proper- ty, and so forth. Discussion in the Nigerian political scene in contemporary 195 times have centred on issues like handling the problem of resource control and Niger-Delta crises, tackling corruption, resolving the issue of the energy crisis and food and personal security. With the 2007 general elections, other major issues in the Nigerian political discourse are electoral reforms and the electoral tribunals’ judgments. Nigerian political discourse generally reflects the idea of a nation on the path to a destination, and the government as one with the task of keeping the nation on this path. Several scholars have focused their research on Nigerian political dis- course, looking at the different aspects. In this work, I shall concentrate my review on the linguistic perspectives on political discourse. Awonuga (2005) examines the linguistic features in the broadcast to the nation by Olusegun Obasanjo, the former president of Nigeria, on August 25, 2002. His analysis reveals a discourse characterised by the use of personal pronouns, coupling, strings of words, analogy, repetition and eight types of metaphor. Another study similar to Awonuga’s, but more restricted in focus, is Adetunji (2006), which also analysed two speeches by Obasanjo, focusing on his use of deix- is. The author was able to demonstrate through these speeches how politi- cians use rhetoric to associate and dissociate themselves from their actions. Opeibi (2006) was a study on negative political advertising in Nigerian newspapers. The author provided a structural and functional description of the emerging trends in negative advertising during political campaigns in Nigeria. In his observation, many of the political office aspirants abandoned positive, issue-focused, image-building advertisements and engaged in rhe- torical strategies of direct attacks on their opponents. His findings are also corroborated by Taiwo (2007), who identified political lampooning of the opposition through newspaper advertisements as one of the major campaign strategies of Nigerian politicians while preparing for the 2007 general elec- tions in the country. Still on political advertising, Opeibi (2007) discusses how Nigerian politicians demonstrate their multilingual creativity. He de- scribes how these politicians engage both the exogenous English and indige- nous languages alongside with the pidginised version of English to drive home their messages. Though the present study shares the same data context with these previous studies on Nigerian political discourse, its focus is dif- ferent in the sense that it is restricted to metaphoric expressions. 4 Perspectives on Metaphors in Political Discourse The study of political language has been the focus of scholars since Classical times. The ancient Greeks were interested in the different possibilities pre- sent in the oratory of politicians (Moreno, 2008: 32). George Orwell drew our attention to political discourse in his satirical novel Nineteen Eighty Four. He portrays a fictional dictatorship in which the old traditional lan- 196 guage Oldspeak is gradually replaced by the new and simplified language Newspeak. Political discourse is an area of study that has attracted scholars from linguistics and political science. Chilton & Schaffner (1997) identify three broad approaches to political discourse analysis, namely: the English, French and German approaches. The Anglophone approaches are mainly studies from the USA and Great Britain, the key figures in these approaches being Norman Fairclough, Roger Fowler, Gunter Kress and Robert Hodge. This approach is based on the functional perspective of language in political discourse. The French scholars consist of a combination of linguists, philos- ophers and political scientists. Their approach relies mainly on applying computer-assisted statistical methods to the political lexicon, establishing a corpus and then making comparisons based on word frequencies (Moreno, 2008: 36). The German approaches focus on text-oriented and action- oriented analyses of the German state within the traditions of pragmatics, sociolinguistics and cognitive linguistics. Some of the most important work specifically on metaphor in political language include: Lakoff & Johnson (1980), Howe (1988), Musolff (2004), Muller (2005), Carver & Pikalo (2008). All these studies underscore the importance of metaphor in the con- struction of political discourse. 5 The Data The data for this study were taken from political speeches of Nigerian politi- cians and Nigerian newspaper publications and commentaries on the coun- try’s politics. In sourcing the data, I examined the political discourse of some key players in the nation since her independence. This presents a rich data from discourses of politicians and political commentators and their concep- tual mappings of issues. My main focus in the data was identification of the metaphors and how the discourses conceptually mapped the targets and do- mains in their metaphoric expressions. 6 Discussion and Findings The analyses reveal that although metaphor is expressed through various means in the Nigerian political discourse, there are three major issues that are commonly focused on in metaphorical expressions. They are: nation, politician and politics. Table 1 below shows the issues and their conceptuali- sation: 197 Table 1. Conceptual Mappings of Targets and Domains in Nigerian Political Metaphors NATION POLITICIAN POLITICS NATION AS FAMILY POLITICIAN AS BUILDER POLITICS AS BATTLE NATION AS PERSON POLITICS AS JOURNEY 6.1 Conceptual Mapping of the Nation Nigeria is a very complex nation in terms of her ethnolinguistic structure. It is not exactly clear just how many languages there are in Nigeria. However, a good estimate may be found in the number of languages listed for Nigeria by Ethnologue, a database of language resources: 521. Politically, Nigeria has been volatile, with ethnic and religious conflicts, and the different groups in the nation struggling for power at the centre. This necessitates the whole idea of the relevance of emphasis on the unity of the nation. Conceptualizing the nation as a family helps politicians to present this message of unity. Ex- amples of the conceptual metaphor of NATION AS FAMILY (with the relevant words in italics) are: (1) We are not to be allowed the selfish luxury of focusing on our homes. (Tafawa Balewa, Independence Day Speech, October 1, 1960) (2) Only a bad father thinks of himself while alive and not of his off- springs. In doing that, such a father will be seen as wicked and irre- sponsible. (Olusegun Obasanjo, Inauguration of the Independent Con- solidating Committee for the Cushioning Measures on High Petroleum Prices, October 11, 2004) (3) To our larger African family, you have our commitment to the goal of African integration. (Umaru Yar’Adua, Inaugural Address May 29, 2007) (4) Our goal is to strengthen the oneness and unity of Nigeria on the basis of the most important asset of our nation—the people. (Olusegun Obasanjo at the opening ceremony of the National Political Reform Conference, February 21, 2005) The conceptual mapping of the nation of Nigeria as a family is seen in the selection of family-related lexicon, in which the nation is perceived as a home (1), the leader as the father, and Nigerians as the offsprings (2), while the continent of Africa is seen as the larger family (3). The government of Olusegun Obasanjo adopted the policy of deregulation of petroleum prices and had to frequently announce increase in the prices of petroleum products. This led to protests and nationwide strikes by the labour unions, which para- lysed economic activities in the nation. The context in which Olusegun 198 Obasanjo made the speech from which example (2) is extracted was that of the inauguration of a committee that would work on how to cushion the ef- fects of the increase in the prices of petroleum products. He had to justify the government’s deregulation of petroleum products prices and allay the fear of Nigerians in the speech. As a “father” to the nation he made it clear in the speech that the policy of deregulation was not an act of wickedness or irre- sponsibility, but that of securing the future of Nigerians. In example (4), the whole idea of the nation being bound together as one family is emphasised. This is necessary in the multilingual/multicultural setting of Nigeria, where whatever leaders do may sometimes be interpreted as ethnic bias. The second conceptual mapping of the nation is that of the NATION AS PER- SON. This is done in order to bring the picture in their messages closer to the people. The nation is perceived as having some human physical attributes such as stature, eyes, a mouth, and so forth. In the context of figurative language, this is called personification. Some examples are: (5) Today, we are embarking on a long and challenging journey to create a united, stable, safe, free and prosperous country. (Atiku Abubakar when announcing his presidential ambitions, November 26, 2006) (6) We must begin to see the Nigerian cup as half full rather than half empty. (Olusegun Obasanjo at the Opening of the National Political Reform Conference) (7) Let us discard the habit of low expectation of ourselves as well as our leaders. (Umaru Yar’Adua, Inaugural Address May 29, 2007) The three politicians from whose speeches the above extracts were taken had different motivations. Atiku Abubakar was the former president of the na- tion, who after his falling-out with the then president, announced his presi- dential ambition. His conceptualisation of the nation as embarking on a chal- lenging journey was used to portray the unsatisfactory level of development of the nation in terms of her unity, stability, safety of lives and property, freedom and prosperity, respectively. His conceptualisation of the nation as a person makes it possible to “create” these issues which he identified in the new nation he was hoping to lead. Olusegun Obasanjo made his speech while inaugurating a national con- ference that would discuss the problems of the nation. His speech focused on what the conference should address and what it should not address. He wished the participants to be constructive and positive in their approach, hence his play on words, encouraging the nation to have a positive visualisa- tion of a cup that is half full rather than half empty. Umaru Yar’Adua was just taking over as the president of the nation and his speech was meant to infuse some hope into the people and encourage them to be positive. This was important for him, as he came into power through an election that was adjudged by international and local independent 199 observers as highly flawed. His choice of the verb discard, which literally means ‘to throw away’, is a reflection of his strong feelings about the Nige- rians’ conception of themselves and their leaders. 6.2 Conceptual Mapping of the Politician The major goals of politicians in making political speeches is to present themselves as people who can be trusted to deliver, meet the people’s aspira- tions and desires and create a stable society for them. Politicians aim at per- suading the electorate and they make use of different kinds of rhetoric strat- egies to achieve these goals. One of the conceptual metaphors used by Nige- rian politicians is that of the POLITICIAN AS BUILDER. This metaphor helps them to present their message of hope to the people, as they present them- selves as foundation-layers, builders and people who restructure and strengthen, for example: (8) We are dedicated to building a viable economy. (Shehu Shagari, Inau- gural Address to the nation, October 1, 1983) (9) This conference is not established to pull down, but to build, it is a gathering to uplift, enhance and strengthen, nurture and cultivate the best, most enduring, most ideal and lasting values that are central to our national growth, development and progress. (Olusegun Obasanjo at the Opening of the National Political Reform Conference) (10) The administration of President Obasanjo has laid the foundation up- on which we can build our future prosperity. (Umaru Yar’Adua, Inau- gural Address May 29, 2007) (11) Thank you for your devoted service, which helped to build up Nigeria into a nation. (Tafawa Balewa, Independence Day Speech, October 1, 1960) The conceptual metaphor of POLITICIAN AS BUILDER is meant to instil some form of confidence in the people they govern. To emphasise the potential in their people, Nigerian politicians portray themselves as builders of the nation and nurturers of the economy. For instance, the government of Olusegun Obasanjo handed over the reins to Umaru Yar’Adua, who is from the same party—the People’s Democratic Party (PDP). Obasanjo had embarked on some reforms during his eight-year rule. His deregulation policy opened up the communications sector and made telephone services available to Nigeri- ans. His privatisation policy brought into the coffers of the country the whopping sum of $7.07 billion, this being the sum generated from the sales of many state-owned companies. He also built Nigeria’s external reserves up to $60 billion before his exit in 2007. He was also able to embark on an anti- corruption crusade, which led to the prosecution of many public office hold- ers. However, his anti-corruption crusade has been criticised by many Nige- 200 rians, as they perceived that it was targeted at his political foes. He would always be remembered as the Nigerian leader who freed his country from the $32 billion debt she owed to the Paris club and also the one who freed his country from the London club. Yar’Adua showcased Obasanjo’s achieve- ments as the foundation upon which he intended to build. The choice of lexicon in the conceptual metaphors of the POLITICIAN AS BUILDER clearly presents the intention of the politicians’ use of language to present this image. For instance, expressions like build, pull down, uplift, nurture, cultivate, lay foundations are used to appeal to the mind-set of the people about their potential to bring about the desired changes. 6.3 Conceptual Mapping of Politics Politics is all about the struggle for power by political actors. In their bid to obtain power, politicians have to align with the electorate and seek their mandate to legitimise their hold on power. Many times, politics is perceived as a battle against the things that are undesirable for the people. To make the people see their passion against such matters, politicians present themselves as people warring against such social ills. It is common in Nigeria to hear politicians use such phrases as “war against corruption”, “war against indis- cipline”, and so forth. This conceptualisation of problems as something to war against presents politicians as the “saviour” or “rescuer” of the people from the social and economic problems of their nation. Some examples are: (12) We will vigorously attack the problem of housing. (Shehu Shagari, Inaugural Address to the nation, October 1, 1983) (13) We will continue to collaborate with fellow African states to reduce conflict and free our people from the leg chains of poverty. (Umaru Yar’Adua, Inaugural Address, May 29, 2007) (14) We have the mandate of Nigerians to guard and protect the interest of Nigerians and to guide and influence the society. (Olusegun Obasanjo at the Opening of the National Political Reform Conference) The government of Shehu Shagari saw housing as a problem that needed to be “attacked”. His choice of the verb and the modifying adverbs shows his perception of the problem. It is not seen as just a problem to be solved, but to be vigorously attacked by the government. Poverty is one of the major prob- lems of most African states and every government in Africa would always have its focus on poverty reduction or total eradication. Yar’Adua, in his inaugural speech, perceived poverty as chains shackling the people’s legs and preventing them from making any progress, and his government and other African leaders as those to rescue the people from poverty. A battle is all about establishing freedom from dominant forces. In Africa, leaders per- ceive national problems as the dominant forces for which they would need 201 the mandate of the people to free them. They also see themselves as those who would free the people from these forces. Obasanjo expressed this in example (14), where he made reference to having the people’s mandate to guard and protect their interest. While politicians see politics as a battle against socio-political ills, politi- cal commentators and the opposition sometimes conceptualise politicians as warring against their people and putting them in shackles. This conceptual mapping is expressed in these two commentaries by political analysts in two different editions of a popular Nigerian newspaper: (15) It is the moment of truth for the Nigerian voter. If he does wish to rid himself from the shackles of politicians, he must plan to be vigilant. (Sonala Olumhense, The Guardian, November 12, 2006) (16) What we must do is to rescue democracy from the enemies of progress and strengthen it. (Reuben Abati, The Guardian, May 28, 2006) The two comments were made as the nation was preparing for the last gen- eral elections of April 2007. The writer conceptualised the Nigerian voters as a people in shackles and admonished them as they prepare for the nation’s general elections to get ready to rescue themselves from the shackles of the politicians. Example (16) conceptualised politicians as the problem of de- mocracy and he urged Nigerians to rescue democracy from the politicians as they prepared to elect another leader. This conceptual metaphor of POLITICS AS BATTLE differs in terms of the focus of the battle for the politicians and political commentators, who are usually government critics. While the poli- ticians focus their battle on the socio-economic problems of the country, the political commentators focus their battle on the politicians, who they per- ceive as the problem in Nigerian politics. The second conceptual mapping is that of POLITICS AS JOURNEY. This conception is formulated to create in the people the sense of better days ahead and allay their fears about developments in their nation. The metaphor of POLITICS AS JOURNEY comes as a very vital way of doing this. It simply presents to the nation the idea of a people on a journey in which the politi- cian is the leader, who promises to take them safely through the journey. The concept of politics as a journey is meant to present the picture that that each step they take on the journey advances them towards a better future. The journey sometimes is conceptualised as a voyage in a ship with the politician as the captain or a journey by road in a car with the politician as the driver: (17) If in the past, the ship of the state became rudderless due to poor han- dling by unskilled captains, you as the present captain decided to go to sleep dreaming that the Lord will steer the ship physically. (Abuba- kar Umar, Open Letter to President Obasanjo) 202 (18) We await it with increasing impatience compelled to watch one coun- try after another overtaking us on the road when we had so nearly reached our goal. (Tafawa Balewa, Independence Day Speech, Octo- ber 1, 1960) (19) We are here to discuss the political path for the nation. (Olusegun Obasanjo at the Opening of the National Political Reform Conference) (20) This day marks an important milestone in our march towards democ- racy. (Umaru Yar’Adua, Inaugural Address, May 29, 2007) (21) We can avoid this monumental setback by toeing the path of reason and engaging in civilized conducts. (The Sun Editorial, October 16, 2006) Example 17 was taken from an open letter to Olusegun Obasanjo as the pres- ident of the nation by one of his critics, who felt he had failed to fulfil his campaign promises. The conceptual metaphor of POLITICS AS JOURNEY here perceived the nation as the ship, past rulers as unskilled captains and Obasanjo as the present captain who had gone to sleep while he believed the Lord will steer the ship. This criticism was targeted at Obasanjo’s religiosity when he was the president. Tafawa Balewa was the first prime minister of Nigeria and he had the privilege of receiving the instruments of the nation’s sovereignty from their British colonial masters on October 1, 1960. The conceptual metaphor he used here presents the British colonies as being on the road and independ- ence as the destination. He saw the nations which got independence before Nigeria as overtaking Nigeria and the day of independence, which he was privileged to oversee as the prime minister, as the goal of the journey. The National Political Reform was perceived by Obasanjo as the political path for Nigeria as she went through her journey to nationhood. The com- mittee would be the one to take the nation through her political journey. The metaphor of the milestone used by Yar’Adua in example (20) conceptualised a marching nation, while example (21) portrays a walking nation, toeing the path of reason. One interesting aspect of the conceptual metaphor of POLI- TICS AS JOURNEY is that the users of the metaphors engaged different vehi- cles to portray the movement of the nation in the journey of politics. The nation is conceptualised as a ship on a voyage and a vehicle on the road. 203 7 Conclusion This study focused on the use of metaphors in Nigerian political discourse. It examined metaphor from the conceptual perspective, describing how politi- cians and political commentators use them as rhetorical strategies to present their message. The study identified three key issues that are the focus of conceptual metaphors in Nigerian political discourse—nation, politician and politics. The nation, which is what politicians struggle to govern, is concep- tualised as a PERSON by giving her some human attributes. This is similar to Lakoff’s (1995) claim that the NATION AS FAMILY metaphor is very common in political discourse. Likewise, Vestermark (2007) identifies the metaphor of the NATION AS PERSON in an analysis of the inaugural speeches of four American presidents. In addition, the present study also identified the con- ceptual metaphor of the POLITICIAN AS BUILDER, describing how Nigerian politicians portray themselves as the builders of the nation. Lastly, politi- cians present the metaphor of POLITICS AS BATTLE and POLITICS AS JOUR- NEY, i.e., a battle against socio-economic problems and a journey towards nationhood, respectively. We also observed that political commentators also conceptualised politics as a battle, but with a different focus of politics as a battle between the politicians and the electorate. References Adetunji, Akinbiyi. 2006. ‘Inclusion and Exclusion in Political Discourse: Deixis in Olusegun Obasanjo’s speeches’, in Journal of Language and Linguistics 5(2): 177–191. Aponte Moreno, Marco. 2008. Metaphors in Hugo Chaveг’s Political Discourse: Conceptualising nation, revolution and opposition. PhD thesis. New York: City University of New York. Awonuga, Chris. 2005. ‘A Stylistic Study of “Sustenance of Democracy” by Nige- ria’s President Olusegun Obasanjo’, in Journal of Social Science, 11(2): 111– 119. Bayley, Paul. 2008. ‘Analysing Language and Politics’, in Mediazione: Rivista online di studi interdisciplinari su lingue e culture. On-line at: http://www.mediazionionline.it/articoli/bayley.htm (retrieved on 15 September 2008). Beard, Adrian. 2000. The Language of Politics. London: Routledge. Bednarek, Monika A. 2005. ‘Construing the World: Conceptual metaphors and event-construal in news stories’, in metaphoric.de 09/2005: 6–32. On-line at: http://www.metaphorik.de/09/bednarek.pdf (retrieved on 15 September 2008). Bosman, Jan. 1987. ‘Persuasive Effects of Political Metaphors’, in Metaphor and Symbol 2(2): 97–113. Carver, Terrell & Jernej Pikalo (eds). 2008. Political Language and Metaphor: In- terpreting and changing the world. London: Routledge. Charteris-Black, Jonathan. 2005. Politicians and Rhetoric: The persuasive power of metaphor. Basingstoke: Palgrave Macmillan. 204 Chilton, Paul & Christina Schäffner. 1997. ‘Discourse and Politics’, in Teun van Djik (ed.), Discourse as Social Interaction. London: Sage. Cienki, Alan. 2005. ‘Researching Conceptual Metaphors that may Underlie Political Discourse’. Paper presented at ECPR Workshop on Metaphor in Political Sci- ence (Granada). On-line at: http://eis.bris.ac.uk/~potfc/Granada/Papers/Cienki.pdf (retrieved on 16 September 2008). Fowler, Roger, Bob Hodge, Günther Kress & Tony Trew. 1979. Language and Control. London: Routledge. Halliday, Michael A. K. 1994. An Introduction to Functional Grammar. London: Edward Arnold. Halliday, Michael A. K. & Robin Fawcett (eds). 1987. New Developments in Sys- temic Linguistics. Vol. 1: Theory and Description. London: Pinter. Howe, Nicholas. 1988. ‘Metaphor in Contemporary American Political Discourse’, in Metaphor and Symbol 3(2): 87–104. Lakoff, George. 1986. ‘A Figure of Thought’, in Metaphor and Symbolic Activity, 1(3): 215–225. Lakoff, George & Mark Johnson. 1980. Metaphors We Live By. Chicago: University of Chicago Press. Linfoot-Ham, Kerry. 2005. ‘The Linguistics of Euphemism: A diachronic study of euphemism formation’, in Journal of Language and Linguistics, 2(2): 227–263. Muller, Ralph. 2005. ‘Creative Metaphors in Political Discourse’, in metaphoric.de 09/2005: 53–73. Online at: http://www.metaphorik.de/09/mueller.pdf (retrieved on 17 September 2008). Musolff, Andreas. 2004. Metaphor and Political Discourse: Analogical reasoning in debates about Europe. Basingstoke: Palgrave Macmillan. Obeng, Samuel G. 1997. ‘Language and Politics: Indirectness in political language”, in Discourse and Society 8(1): 49–83. Opeibi, Tunde. 2007. ‘One Message, Many Tongues: An exploration of media mul- tilingualism in Nigerian political discourse’, in Journal of Language and Poli- tics 6(2): 223–248. — 2006. ‘Political Marketing or Political “Macheting”? A Study of Negative Cam- paigning in Nigerian Political Discourse’, in Reproduction and Innovation in Language and Communication in Different Language Cultures 16. On-line at: http://www.inst.at/trans/16Nr/01_4/opeibi16.htm (retrieved on 17 September 2008). Taiwo, Rotimi. 2007. ‘Satirising Politicians and Public Officers in Nigerian News- papers’, in The International Journal of Language, Society and Culture 22(3): 19–28. Threadgold, Terrry. 2003. ‘Cultural Studies, Critical Theory and Critical Discourse Analysis: Histories, remembering and futures’, in Linguistik Online 14: 5–37. Vestermark, Ida. 2007. ‘Metaphors in Politics: A study of the metaphorical personi- fication of America in political discourse’. C-Extended Essay, Lulea University of Technology and Culture. Online at: http://epubl.ltu.se/1402- 1773/2007/080/LTU-CUPP-07080-SE.pdf (retrieved on 16 August 2008). Warren, Beatrice. 1992. ‘What Euphemisms Tell us about the Interpretation of Words’, in Studia Linguistica 46(2): 128–172. Wilson, John. 2001. ‘Political Discourse’, in Schiffrin, Deborah, Deborah Tannen, and Heidi E. Hamilton (eds), The Handbook of Discourse Analysis. Oxford: Blackwell Publishing, 398–415. Yusuf, Yisa, K. 2003. ‘Dysphemism in the Language of Nigeria’s President Olu- segun Obasanjo’, in AILA Review 16: 104–119. 205 206 Piotr Twardzisz Metaphors in Commercial Contracts Abstract. The present paper analyses some non-obvious traces of figurative mecha- nisms in an instance of language for specific purposes, namely language used in legal contexts. Naturally, the material from the field of legal language has had to be tailored to suit the scope and purpose of this paper. Selected examples of fragments of contract clauses from English legal language have been analysed and several metaphors have been elicited. In a world described by means of legal language, companies are living beings which occasionally may get injured by other living beings. Difficult or problematic situations are like knots which are to be untied. Ideas like rights and duties are physical objects to be carried from one location to another. The proposed study should constitute a contribution to research on the use of figurative language in specialist communication, particularly in its written ver- sion. Keywords: figurative language, metaphor, legal language, language for specific purposes (LSP), language for general purposes (LGP) 1 Language for specific purposes The last two decades or so have seen the need for yet another division within linguistics, or applied linguistics, to be more precise. The dichotomy in ques- tion is the one between language for general purposes (LGP) and language for specific purposes (LSP). Artificial though the division may be felt, it has become firmly established within applied linguistics. As the present article concerns a further sub-division within LSP, namely legal language, let us briefly review some of the major tenets of the distinction in question, with necessary shortcuts. A somewhat secondary nature of specialist language, in comparison to general language, has been observed, for example, in Lukszyn (2007). Ac- cording to Grucza (2006: 34), specialist languages ought to be studied as autonomous languages. At the same time he notes that specialist languages are not complete languages, in the linguistic sense of the word language, and are closely tied to some ‘general’ or ‘basic’ language. According to Grucza, the phonetics and grammars of specialist languages overlap those of their respective general languages fairly completely. However, their lexical stocks and textual patterns overlap only partially. He stresses the fact that specialist 207 languages are not just any variants of general languages, because one cannot use either specialist language or general language with reference to the same scope of reality. Therefore, he contends, specialist languages are specific human languages created by specialists for purposes of professional commu- nication within their respective professional communities. It should come as no surprise that it is terminology, rather than anything else, that constitutes the main target of LSP research. Relatively little atten- tion has been given to the role of figurative language, and metaphor in par- ticular, within LSP. However, in recent years, there have been some attempts to detect metaphor in specialist discourse (see, for example, Charteris-Black 2000, Charteris-Black & Ennis 2001, Skorczyńska 2001, White & Herrera 2003, and others). By definition, language used by professionals, say medi- cal doctors, lawyers, IT specialists, or language used in the literature pertain- ing to their respective fields is meant to be unambiguous and as precise as possible. This runs counter to what we observe in regular, everyday speech or literary language, where metaphor seems to be the norm. 2 Legal language In calling the target of my research legal language, I am taking a shortcut overlooking the distinction typically made between legal language and lan- guage of law, where the former often refers to language used when talking about legal matters, and the latter typically goes to the written language of legal documents, contracts, statutes and so on. There is no certainty whether legal language can be identified as (a) lan- guage, in any sense of the term. In his highly acclaimed book Legal Lan- guage, which in fact discusses legal English, Peter Tiersma asks the ques- tion: “what is legal language exactly? A separate language? A dialect? Jar- gon? Argot? A style?” (1999: 142). The author points out that none of the above can be thought of as sufficient, if taken separately. The very term lan- guage may be deemed inappropriate, especially when one takes away from legal English all its distinctive features: leaving out features characteristic of general English, one is left with something incomplete, unable to function as a language per se. For Tiersma, legal language is not a dialect, which is geo- graphically motivated, a jargon, which is related to vocabulary, an argot, which has overtones of secretiveness, or a style, as features of legal language go beyond those typically thought of as stylistic in nature. A somewhat more promising terminological candidate is the notion of sublanguage, in that many of its features, in terms of syntax, morphology and semantics, seem to apply to legal language. Abstracting away from Tiersma’s remarks, an obvi- ous reservation that one may have with reference to the notion of sub- language is the semantic contribution of the prefix sub-, which often carries 208 a pejorative sense, presumably unwelcome in the case of the collocation legal sublanguage. Owing to the fact that the debate over the precise linguistic status of legal language is largely unresolved, in this paper I resort to the notion language. As establishing an undisputed and indisputable label for the target of my research is not a priority here, I bypass this controversy and focus on the characteristic features of legalese, with particular stress on its metaphoric nature. 3 Characteristic features of legal language When talking of legal language, in particular legal English, one needs to keep in mind the number of varieties of English, as well as the multitude of legal systems which are practised through those varieties of English. In view of the above, one may wonder whether a given variety of English, say Amer- ican English, is reserved for the American legal system(s) solely. There does not seem to be complete agreement over this issue. On the one hand, as de Groot (1998: 21) maintains, the language of the law, being a system-bound language, is related to a specific legal system. On the other hand, however, as pointed out in Kocbek (2006: 235), one legal system may use different legal languages, while one language territory may be shared by different legal systems (e.g. the UK or the USA). Undoubtedly, there are features that are often found to be typical of legal language. Among them there are lengthy and complex sentences, which means anything from 10 to 179 words per sentence, according to one study referred to in Tiersma (1999: 56). Apart from length, sentences in legal texts are known for their complexity, which is suggestive of multiple imbedded and conjoined clauses. Legal language, specifically in its written form, is notorious for wordiness and redundancy (e.g. last will and testament, in the event that, during the same time that, etc.), which is intended to remove any possibility of ambiguity. Largely in the same vein, though without much reason, one also finds in legal written language numerous conjoined phrases and lists of words (e.g. in accordance with and subject to, give, devise and bequeath, null and void, cease and desist, etc.). Last but not least, legal doc- uments are full of unusual sentence structures, which frequently, though not exclusively, go back to word-by-word translations from Latin. In their search for precision, or in their efforts to avoid ambiguity, legal document drafters favour repetition of the noun phrase instead of the use of pronouns, as in ... the Buyer shall exercise the Buвer’s right to ... instead of ... his right to .... On the contrary, the prevalent use of the pronoun he stands in opposition to the apparent drive toward avoiding ambiguity, as it also includes she and it. A similar ambiguity may be found in the use of the pre- 209 sent tense, for instance, The Company expresses consent for …, which can equally well be interpreted as referring to the past ... expressed ... and the future ... will express .... Lack of precision may also be found in numerous instances of passive voice sentences, where the agent can remain unmen- tioned, as in Share capital can be reduced …, though this technique is not very common in contracts, where it is vital to spell out who should do what. Also, legal language cannot be said to be semantically unambiguous owing to the plethora of words, in many cases adjectives, which contribute an am- biguous element, as in reasonable care, due process, indecent language, to name just a few. The same is true of verbs and nouns, as exemplified by to harass, harassment, pornography, restraining orders, etc., which in many cases have to be further and specifically defined. Bypassing a whole gamut of other features characteristic of legal lan- guage, let us proceed to the discussion of its metaphorical aspect. 4 Metaphors in legal English Almost three decades of intense research have shown that metaphor is not a mere decorative device, previously thought of as belonging mostly to literary and rhetorical language, but a complex conceptual phenomenon pervading the way we think and speak. It is both thought and language that the notion of metaphor applies to. Being fundamentally a conceptual phenomenon (thought metaphor), metaphor reveals itself through concrete language ex- pressions (language metaphor). Research in the area of metaphor has con- centrated on conceptual metaphor, that is, the mental construct that language users do not actually come up with, and that researchers reconstruct as a possible thought process governing what is actually said or heard. Numerous researchers in the field of cognitive linguistics, going back to seminal studies in metaphor such as Lakoff & Johnson (1980), have pointed out that ordinary speakers regularly use linguistic metaphors in everyday speech and those metaphors reveal underlying thought processes, that is conceptual metaphors (cf. Kövecses 2010, Steen 2009, and others). In fact, metaphor is claimed to be omnipresent, rather than peripheral or marginal, and our conceptual system is metaphorical in nature. Research into metaphor has largely concentrated on language for general purposes, using our earlier terminology. Metaphor in language for specific purposes has not been a very popular area of research, with some exceptions (see section 1, above). Resistance to accepting the presence of metaphor in thinking about legal language is more of an accepted standard in linguistic research than an exception. The two seminal books in the field of language and law, namely Mellinkoff (1963) and Tiersma (1999), though exhaustively and thoroughly covering the interface between law and language, do not 210 seem to attach any importance to the phenomenon of metaphor in legal lan- guage. Tiersma (1999: 128) stresses the significance of the autonomous na- ture of written legal language, as the writer and the reader cannot be as- sumed to share the same background information or knowledge. Therefore, according to Tiersma (1999: 128), “[...] because of the seriousness of the topic, we can safely assume that humor, irony, figurative usage, and similar literary devices will be avoided”. While leaving humour and irony out of legal written language seems de- sirable for a number of reasons, it remains to be seen whether avoiding fig- urative language in legal documents is at all possible. Several authors have noted that metaphor is indeed an inalienable part of legal language. Hibbitts (1994) examines the contemporary shift from visually-oriented to aurally- oriented figures of speech in American legal language. We frequently use the following expressions: we observe the law, in the eyes of the law, high courts review the decisions of inferior tribunals, etc., as though we consid- ered law as a matter of looking. Also, we often talk of law as if it were some- thing to look at, for instance, a text, a body or a structure. Visual metaphors can be detected by giving law the visual quality of hue, when one makes a property claim under colour of title, discourages yellow dog contracts, ad- heres to black letter rules or makes securities trading subject to blue sky laws (also, see, for example, Winter 2008). Hibbitts further observes that there is a tendency, among feminist, African-American, Hispanic and Jewish legal scholars, to employ aural metaphors in their writings. For some of those scholars, law is a matter of voice, speaking and even, occasionally, of sing- ing. In his paper ‘The Myth of Corporate Veil Protection: Are Your Assets at Risk?’, Graham (2002) describes the corporate veil as a legal metaphor which stands for “[...] the liability shield that a corporation or a limited lia- bility company is supposed to create around its owners and their personal assets”. Whenever a court pierces the corporate veil, it disregards the exist- ence of a higher form of business organization, since, for example, the own- ers of a corporation failed to maintain some corporate requirements and for- malities. If so, the business is no longer viewed as a corporation but as a sole proprietorship or general partnership and the owners’ personal assets are no longer immune to litigation. A serious treatment of metaphor in the context of legal language is pre- sented in Schane (2006). Apart from pointing out expressions taken from legal discourse which originated as metaphors, for instance, a meeting of the minds, a ripening of obligations, a binding agreement, a broken contract, etc., Schane acknowledges the need for another kind of metaphor –– the ‘legal fiction’. An example of that may be the well-known ‘the corporation as a person’ fiction, which will be discussed later in this paper. As compa- nies or corporations can, in the legal sense, be involved in many activities typically thought of as reserved for humans, such as owning property or 211 entering into agreements, it should come as no surprise that many legal sys- tems often deal with corporations as if they were indeed persons. The current interest in specialist, or at least non-literary, texts is observed in Stålhammar (2006), who explores the so-called grammatical metaphor and metonymy in EU documents. Having acknowledged the increasing global need for English for special purposes, Stålhammar goes on to distin- guish grammatical metaphor from semantic metaphor. The latter is defined as the substitution of one word by another, while the former consists in “a substitution of one grammatical class, or one grammatical structure, by an- other”1. More precisely, grammatical metaphor is found to stand behind the reification of processes into objects, which not only alters the grammar of the text but also the reader’s reaction to it. Stålhammar (2006: 100) points out that grammatical metaphor, owing to its condensing function, “[...] is a very economical means of packaging information and is consequently fre- quently used in scientific and technical information”. With this in mind, the author proceeds to investigate a particular type of grammatical metaphor, called in her paper grammatical metonymy, in two different versions of the proposed constitution of the European Union. From the above sketch, one can glean that metaphor is not a non-existent phenomenon in specialist (legal) texts and it has attracted the attention of a number of scholars. Let us now proceed to our analysis of a selection of contract clauses, with the aim to establish metaphoric patterns operating at the conceptual level. 5 Metaphors in commercial contracts Naturally, the broad and diverse field of legal language has had to be limited for the scope of this short paper. The commercial contract has been chosen as the genre of texts analysed in this study. Approximately 60 contracts and agreements in the English language have been selected and analysed. All of these texts have been taken from Bogudziński et al. (2000), which is a bilin- gual English-Polish collection of relevant source texts. The actual documents differ in size from one page to several pages in length. The material is not tagged, and manual search is the only reliable search method (cf. Steen 2009). All the fragments from (1) to (40) below come from the above- mentioned source. Also, it needs to be noted that the fragments in question are not unique in any way and they can be found in many different commer- cial contracts and agreements. In order to achieve maximum transparency, only the most relevant fragments have been copied and the remaining part of each sentence, irrelevant for this study, has been deleted. Additionally, the 1 The quote originally goes back to Halliday & Martin (1993: 79). 212 pivotal part of each fragment has been underlined. By the pivotal part is meant a word or a phrase in which a given metaphor seems to ‘reside’. Throughout the rest of this paper, the view expressed in Lakoff & John- son (1980: 5) will be subscribed to, namely that “[t]he essence of metaphor is understanding and experiencing one kind of thing in terms of another”. I also rely on the convenient distinction between the source and target do- mains and an intricate mapping of the relations between them (cf. Kövecses 2010, Ruiz de Mendoza Ibáñez 2000, and others). In general, the conceptual metaphors which we will consider below have a structure represented as: X IS Y or X DOES Y. That is because “[...] most of our normal conceptual system is metaphorically structured; that is, most concepts are partially understood in terms of other concepts” (Lakoff & Johnson 1980: 56). In what follows, let us establish metaphorical patterns appearing in legal contexts, particularly, in commercial contracts. 5.1 The COMPANY IS A PERSON metaphor As repeatedly stated by Lakoff & Johnson (cf. Lakoff & Johnson 1999: 3–4), “[t]he mind is inherently embodied. Thought is mostly unconscious. Abstract concepts are largely metaphorical” and “[r]eason [...] arises from the nature of our brains, bodies, and bodily experience”. One of the most common forms of business activity, namely a company, is frequently thought of as a person. This conception is omnipresent in all kinds of legal writings, includ- ing contracts. There is also the term legal person applied, among other legal entities, to a limited liability company (LLC). One way to look at an LLC is to assume the legal perspective according to which an LLC satisfies a num- ber of characteristics of a (legal) person, for instance, formation, bearing liability, carrying out various duties, termination, etc. A different way to look at an LLC is from the conceptual point of view, something which we, as humans, owing to our embodied minds, are particularly well equipped to do. If our conceptual systems grow out of our bodies and meaning is grounded in and through our bodies, then our minds are indeed embodied. Granted that our minds are embodied, entities such as companies, with which humans interact on a daily basis, are themselves conceived of as em- bodiments. The conception of interaction between people and companies will be significantly facilitated, provided companies are conceived of as human beings. Similarly to people, especially active ones, companies can operate, create and close entities, appoint people, and so on, as in: (1) The Company can operate in the Czech Republic and abroad. (2) The Company can create and close branches. (3) The Company appoints the following persons for the day to day con- tacts with the Legal Adviser. 213 People are prototypical owners and possessors of goods and objects. If the metaphor in question is in operation, it should be natural to assume that companies (that are humans) also possess goods and objects, as shown in (4): (4) The Company does not have any arrears in its financial liabilities. Communication is one of the most fundamental abilities and skills that hu- man beings are endowed with. And so are companies, conceived of as peo- ple, as is seen below: (5) The Company expresses consent for […]. (6) The Company can appeal directly to […]. (7) The Legal Adviser is obliged to inform the Company of […]. (8) The Company shall be obliged to provide the Legal Adviser with the information […]. Companies, like people, enjoy certain rights: (9) The Company is entitled to demand […]. Similarly to people, companies show interest: (10) The Company states that it is interested in purchasing undeveloped land. As seen in the above examples, companies exhibit numerous traits common- ly thought of as human-like. However, it needs to be clearly stated that there is an intrinsic asymmetry between the two participants of this metaphor, namely a company and a person. While a company can be conceived of as a person, the opposite relation is not possible. A company is part of the target domain, which, by definition, is more abstract, but is related via correspond- ences to the source domain, with its more tangible element, namely a person. 5.2 The ENTITY IS A PERSON metaphor As Lakoff & Johnson (1999: 6) point out, “[b]ecause our conceptual systems grow out of our bodies, meaning is grounded in and through our bodies”. In the previous section we have seen that companies exhibit human-like fea- tures. Now, let us examine how less tangible entities are conceptually related to people. We are advised in Lakoff & Johnson (1999: 34) that “[w]e project fronts and backs onto objects. […] The concepts front and back are body- based. They make sense only for beings with fronts and backs”. Legal writ- 214 ten language abounds in statements making reference to fronts in particular, for instance in front of all bodies, institutions and offices. If so, entities such as institutions and offices are viewed as consisting of fronts, which, in a way, implies the existence of backs. Different kinds of documents can be conceived of as consisting of body- parts, such as arms, which can be used to cover something: (11) [...] a power of attorney to sell the Item, which covers […] (12) The invoice covering the remuneration for preparing the analysis […]. An entity, such as property, can be thought of as a free being, similar to a human being: (13) The Property is free of any encumbrances. Things and physical objects, according to ontological metaphors, are further specified as being persons, which is highlighted in Lakoff & Johnson (1980: 33): (14) The Seller declares that the vehicle defined in Art. 1 above is free of all legal defects and third party rights. The frequent Force Majeure clause is construed in such a way as if the po- tential cause of a misfortune were an active human being: (15) […] as a result of an act of Force Majeure […] (16) […] the duration of this act of Force Majeure […] Nevertheless Force Majeure is customarily defined as referring to any exter- nal event of an extraordinary nature, such as riots, strikes, collective dis- putes, armed conflicts and natural disasters. None of these make direct refer- ence to a human being. 5.3 The COMPANY IS A TEMPORAL BEING metaphor Somewhat similarly to the COMPANY IS A PERSON metaphor, a company can be conceived of as any entity, not necessarily a human being, whose ‘life’ is determined at both ends. Viewed as such, companies are created at some stage: (17) […] they are forming a limited liability company […] (18) The Partners have decided to establish a limited liability company […]. (19) The Partners hereby establish a civil partnership [...]. 215 Unsurprisingly, a company’s life comes to an end in the course of time: (20) The duration of the Company is unlimited. (21) Winding-up of the Company is preceded by its liquidation. This metaphor is very similar to the one discussed in section in 5.1, above. The difference between the two is that in the COMPANY IS A TEMPORAL BE- ING metaphor there is no direct reference to a human being, though such reference cannot be precluded altogether. Here a company is put in corre- spondence with any living organism, whose life is naturally limited. 5.4 The NON-OBJECT IS AN OBJECT metaphor It is presumably our natural tendency to construe a less concrete concept in terms of a tangible object, one with a clear shape that we are familiar with through our experience. That experience with concrete objects constitutes the basis for viewing events, activities and ideas as more substantial beings (cf. Lakoff & Johnson 1980: 25, 112). Let us analyse several instances of the metaphor in question. An activity, state, property, right and duty can be conceived of as an object that can be carried and moved around by people: (22) I declare that I am conducting economic activity […]. (23) […] the transfer of ownership of the Property to the Company. (24) In the event of any circumstances occurring at AAA preventing the submission to the Company of the Property according to the condi- tions […]. (25) The Seller hereby transfers to the Buyer the right to dividends. (26) The Agent cannot transfer the rights and duties from this Contract to third parties. (27) In the event that some Shareholders choose to waive this right then it passes to the remaining Shareholders […]. (28) All risks connected with destruction, damage or loss of the Item for Sale shall be transferred to the Buyer upon collection of the Item for Sale. We also tend to view prices and amounts of money as vertically oriented objects or points on vertically oriented scales that can be raised or lowered: (29) […] the sale price for the Item cannot be less than […] (30) The Company’s share capital can be increased without amending the Articles of Association increasing the nominal value of the existing shares. 216 (31) The place, amount and date for payments for the new shares […]. (32) Share capital can be reduced […]. The abstract concept of cost is structured in terms of the tangible notion of a heavy object or even a burden, as in: (33) The costs of preparing these Articles of Association shall be borne by the Partners in equal proportion. Intangible concepts such as shares can be seen as tangible objects that can be held. (34) Each Shareholder can hold more than one share. Viewing all the above abstract concepts as concrete objects that can be han- dled, held, passed around and so on, allows us to not only refer to them, but also to discuss their role in all kinds of scenarios. 5.5 The NON-CONTAINER IS A CONTAINER metaphor The idea of viewing people and our visual field as containers probably goes back to Lakoff & Johnson (1980: 29–30). Also, there are some other publi- cations in which the metaphorical container has been proposed in other spe- cialist contexts (see, for example, Charteris-Black 2006, Drulák 2006, Chil- ton 1996, and others). Moreover, legal written contexts provide us with nu- merous examples in which intangible concepts are conceived of as contain- ers with bounding surfaces and in-out orientations. The following are examples of abstract notions such as good faith, issue and voting, which are all viewed as container-like objects, as in: (35) The parties shall agree in good faith to the remaining conditions. (36) The provisions of the Civil Code and Commercial Code shall be ap- plicable in issues not governed by this Memorandum of Understand- ing. (37) The Management Board passes resolutions in open voting. Objects such as different kinds of documents can also be thought of as con- tainers for words: (38) The minutes should contain the place and date […]. (39) [...] the activities foreseen in the Commercial Code [...] (40) […] in the Statute […] 217 It should be remembered, though, that the semantic contribution of the prep- osition in or the verb contain cannot be ignored, either. It may be argued then that it is not the alleged container-like character of the nouns corre- sponding to the objects of in above, but the notion of inclusion present in the semantics of the atemporal relation IN, using the terminology elaborated in Langacker (1987, 1991). As pointed out in Langacker (1987: 226), “[b]asic conceptual relations are fully schematic in their domain specifications”. Al- though one may describe the above sentences without evoking spatial meta- phors, the contribution of such a metaphor cannot be ruled out. So, on the one hand, the relation in itself introduces the concept of inclusion and re- quires that the object of in be viewed as some kind of container, albeit an abstract one. On the other hand, the noun corresponding to the object of in is metaphorically structured as a container; given that it follows in, the concept of a container-like relation is additionally highlighted. Presumably the issue will remain unresolved, as both options seem equally plausible. 6 Conclusion The present paper provides evidence for the existence of metaphor in legal written language. This variety of language for specific purposes is full of terms which designate abstract notions, most of which are not directly acces- sible, as they do not denote tangible referents. To partially overcome this hindrance, legal language employs metaphor as a convenient tool which makes the comprehension of otherwise abstract concepts easier. To reduce the apparent deficit of concepts directly understood, legal language resorts to metaphorical expressions that provide some means of processing these inac- cessible or less accessible notions. It is relatively effortless to conceive of a company as a (legal) person instead of something as elusive as a business organisation that makes or sells goods or services. This conceptual metaphor does not seem to be reserved for English only. Similar metaphors, based on data found in the Polish legal language, have been proposed in Twardzisz (2011). Ease of comprehension may not be the ultimate or sole cause of metaphor formation in the case of legal language. As we are regularly reminded by Lakoff & Johnson (1980), most of our normal conceptual system is thor- oughly metaphorical and it is thus the case that most concepts are partially understood in terms of other concepts. Therefore, if our normal conceptual system is metaphorical, the arbitrary dichotomy between LGP and LSP should not be understood as a borderline for the use of metaphor. With met- aphor permeating our entire language capacity, there is no reason why LSP should be left immune to it. In fact, it would be odd if metaphor did not 218 make itself present in more specialised contexts, as they need a source for the vocabulary that one uses in talking about intangible experience. References Bogudziński, Jacek, Konrad Buczkowski & Andrzej Kaznowski. 2000. Compendi- um of Legal Templates. Warsaw: C. H. Beck. Charteris-Black, Jonathan. 2006. ‘Britain as a Container: Immigration metaphors in the 2005 election campaign’, in Discourse & Society 17(5): 563–581. —. 2000. ‘Metaphor and Vocabulary Teaching in ESP Economics’, in English for Specific Purposes 19(2): 149–165. Charteris-Black, Jonathan & Timothy Ennis. 2001. ‘A Comparative Study of Meta- phor in Spanish and English Financial Reporting’, in English for Specific Pur- poses 20(3): 249–266. Chilton, Paul A. 1996. Security Metaphors: Cold War discourse from Containment to Common House. Bern: Peter Lang. De Groot, Gerard-René. 1998. ‘Language and Law’, in Hondius, Ewoud (ed.), Neth- erlands Report to the Fifteenth International Congress of Comparative Law. Antwerp: Intersentia, 21–32. Drulák, Petr. 2006. ‘Motion, Container and Equilibrium: Metaphors in the discourse about European integration’, in European Journal of International Relations 12(4): 499–531. Fuller, Lon L. 1967. Legal Fictions. Stanford, CA: Stanford University Press. Graham, Stanford A. 2002. ‘The Myth of Corporate Veil Protection: Are your assets at risk?’. On-line at: http://bulletproofveil.com/pdf/whitepaper.pdf (retrieved on 23 July 2008). Grucza, Sambor. 2006. ‘Idiolekt Specjalistyczny — Idiokultura Specjalistyczna — Interkulturowość Specjalistyczna’ [‘Specialist Idiolect — Specialist Idioculture — Specialist Interculture’], in Lewandowski, Jan et al. (eds), Jęгвki Specjalistyczne 6. Warsaw: University of Warsaw, 30–49. Halliday, M. A. K. & James R. Martin. 1993. Writing Science: Literacy and discur- sive power. London: Falmer Press. Hibbitts, Bernard J. 1994. ‘Making Sense of Metaphors. Visuality, Aurality, and the Reconfiguration of American Legal Discourse’. Online at: http://faculty.law.pitt.edu/hibbitts/meta_int.htm (retrieved on 5 July 2008). Kocbek, Alenka. 2006. ‘Language and Culture in International Legal Communica- tion’, in Managing Global Transitions 4(3): 231–247. Kövecses, Zoltán. 2010. Metaphor: A Practical Introduction. 2nd ed. Oxford: Oxford University Press. Lakoff, George & Mark Johnson. 1999. Philosophy in the Flesh. The Embodied Mind and its Challenge to Western Thought. New York: Basic Books. ––. 1980. Metaphors We Live By. Chicago: The University of Chicago Press. Langacker, Ronald W. 1991. Foundations of Cognitive Grammar. Vol. 2: Descrip- tive Application. Stanford: Stanford University Press. ––. 1987. Foundations of Cognitive Grammar. Vol. 1: Theoretical Prerequisites. Stanford: Stanford University Press. Lukszyn, Jerzy. 2007. ‘Tekst Specjalistyczny pod Lingwistyczną Lupą’ [‘Specialist Text through a Linguistic Magnifying Glass’], in Kornacka, Małgorzata (ed.), Jęгвki Specjalistyczne 7. Warsaw: University of Warsaw, 51–70. 219 Mellinkoff, David. 1963. The Language of the Law. Boston: Little, Brown & Co. Ruiz de Mendoza Ibáñez, Francisco José. 2000. ‘The Role of Mappings and Do- mains in Understanding Metonymy’, in Barcelona, Antonio (ed.) Metaphor and Metonymy at the Crossroads. A Cognitive Perspective. Berlin: Mouton de Gruy- ter, 109–132. Schane, Sanford. 2006. Language and the Law. London: Continuum. Skorczyńska, Hanna. 2001. ‘Metaphor in Scientific Business Journals and Business Periodicals: An example of the scientific discourse popularisation’, in Ibérica 3: 43–60. Stålhammar, Mall. 2006. ‘Grammatical Metaphor/Metonymy in the Treaty Estab- lishing a Constitution for Europe: A comparison between the English and Swe- dish versions’, in Nordic Journal of English Studies 5(1): 99–117. Steen, Gerard J. 2009. Finding Metaphor in Grammar and Usage: A methodological analysis of theory and research. Amsterdam: John Benjamins. Tiersma, Peter. 1999. Legal Language. Chicago: University of Chicago Press. Twardzisz, Piotr. 2011. ‘Metaphorical Expressions in Legal Language: Evidence from Polish’, in Grygiel, Marcin & Laura A. Janda (eds), Slavic Linguistics in a Cognitive Framework. Bern: Peter Lang, 317–331. White, Michael & Honesto Herrera. 2003. ‘Metaphor and Ideology in the Press Coverage of Telecom Corporate Consolidations’, in Dirven, René, Roslyn M. Frank & Martin Pütz (eds), Cognitive Models in Language and Thought. Ideol- ogy, Metaphors and Meanings. Berlin: Mouton de Gruyter, 277–323. Winter, Steven L. 2008. ‘What is the “Color” of Law?’, in Gibbs, Raymond W., Jr. (ed.), The Cambridge Handbook of Metaphor and Thought. Cambridge: Cam- bridge University Press, 363–379. 220 List of contributors Ludmilla A’Beckett is an Adjunct Research Fellow at the School of Lan- guages, Cultures and Linguistics, Monash University, Australia. Email: lyudmyla.antypenko-a'beckett@monash.edu Christina Alm-Arvius was an Associate Professor at the Department of Eng- lish, Stockholm University. Antonio Barcelona is a Professor of English Semantics and Grammar at the Department of English and German, University of Córdoba, Spain. Email: ff1basaa@uco.es Bárbara Eizaga Rebollar is a Professor at the University of Cádiz, Spain. Email: barbara.eizaga@uca.es Shelley Ching-yu Hsieh is a professor at Department of Foreign Languages and Literature in National Cheng Kung University, Taiwan, and a Chair professor at Leiden University, The Netherlands. Email: ching- yu2@gmail.com Elena Kolodkina has retired from her work at the Applied English Depart- ment, Southern Taiwan University of Technology. Email: elena.kolodkina@gmail.com Anna Kryvenko is a Lecturer at Kyiv National Linguistic University, Ukraine. Email: annakry@fulbrightmail.org Ana Roldán-Riejos is an Associate Professor at the Technical University of Madrid (UPM), Spain. Email: aroldan@caminos.upm.es Gerard Steen is Professor of Language and Communication at VU Universi- ty, Amsterdam, The Netherlands. Email: g.j.steen@vu.nl Rotimi Taiwo is a Senior Lecturer at the Department of English, Obafemi Awolowo University, Nigeria. Email: ferotai@yahoo.com. Website: www.rotimitaiwo.net Piotr Twardzisz is a Lecturer in the Faculty of Applied Linguistics, Universi- ty of Warsaw, Poland. Email: p.twardzisz@uw.edu.pl 221 STOCKHOLM STUDIES IN ENGLISH Published by the University of Stockholm. Founded by Arvid Gabrielson. Editors: Kingsley Bolton, Harald Fawkner, Nils-Lennart Johannesson, Magnus Ljung, Gunnel Melchers. Orders for single volumes can be addressed directly to the distributor: Stockholm University Library SE-106 91 Stockholm Phone: + 46 (0)8-16 29 52 E-mail: acta@sub.su.se www.sub.su.se I. Axel Wijk, The Orthography and Pronunci- VIII. Conrad Lindberg, MS. Bodley 959. ation of Henry Machyn, the London Diarist: Genesis–Baruch 3.20 in the Earlier Version A Study of the South-East Yorkshire Dialect of the Wycliffite Bible. Volume 2: Leviti- in the Early 16th Century. Uppsala, 1937. cus–Judges 7.13. Stockholm, 1961. Pp. 281. Pp. xi+299, with a loose leaf of Additions IX. Sverker Brorström, The Increasing Fre- and Corrections, and two plates. quency of the Preposition About during the II. Arthur Ahlgren, On the Use of the Defi- modern English Period, with Special Refer- nite Article with ‘Nouns of Possession’ in ence to the Verbs Say, Tell, Talk, and Speak. English. Uppsala, 1946. Pp. xv+221. Stockholm, 1963. Pp. 335. III. Bror Danielsson, Studies on the Accen- X. Conrad Lindberg, MS. Bodley 959. tuation of Polysyllabic Latin, Greek, and Genesis–Baruch 3.20 in the Earlier Version Romance Loanwords in English, with Spe- of the Wycliffite Bible. Volume 3: Judges cial Reference to Those Ending 7.13–II Paralipomenon. Stockholm, 1963. Pp. in -able, -ate, -ator, -ible, -ic, -ical and -ize. 395. Stockholm, 1948. Pp. xvi+644. XI. Bror Danielsson, John Hart’s Works on IV. Sven L. Fristedt, The Wycliffe Bible. English Orthography and Pronunciation Part I. The Principal Problems connected (1551, 1569, 1570). Part II. Phonology. with Forshall and Madden’s Edition. Stock- Stockholm, 1963. Pp. 294. holm, 1953. Pp. xvi+148; seventeen plates. XII. Bror Danielsson, Sir Thomas Smith. V. Bror Danielsson, John Hart’s Works on Literary and Linguistic Works (1542, 1549, English Orthography and Pronunciation 1568). Part I. Certaigne Psalmes or Songues (1551, 1569, 1570). Part I. Biographical and of David. Stockholm, 1963. Pp. 44. Bibliographical Introductions, Texts and XIII. Conrad Lindberg, MS. Bodley 959. Index Verborum. Stockholm, 1955. Pp. 338. Genesis–Baruch 3.20 in the Earlier Version VI. Conrad Lindberg, MS. Bodley 959. of the Wycliffite Bible. Volume 4. I Esdras– Genesis–Baruch 3.20 in the Earlier Version of Ecclesiasticus 48.6. Stockholm, 1965. Pp. the Wycliffite Bible. Volume 1: Genesis & 462. Exodus. Stockholm, 1959. Pp. 213. XIV. Sverker Brorström, Studies on the Use VII. Axel Wijk, Regularized English. An of the Preposition Of in 15th Century Corre- Investigation into the English Spelling Re- spondence, with Special Reference to Con- form Problem with a New, Detailed Plan for structions Differing from Present-day Usage. a Possible Solution. Stockholm, 1959. Pp. Stockholm, 1965. Pp. 74. 361. XV. Lars-G. Hallander, Old English Verbs nomia Anglica (1619). Part II. Biographical in -sian. A Semantic and Derivational Study. and Bibliographical Introductions. Transla- Stockholm, 1966. Pp. 619. tion and Notes. Stockholm, 1972. Pp. 245. XVI. Sven Jacobson, Unorthodox Spelling XXVIII. Sven L. Fristedt, The Wycliffe in American Trademarks. Stockholm, 1966. Bible, Part III. Relationships of Trevisa and Pp. 53. the Spanish Medieval Bibles. Stockholm, XVII. Bror Danielsson and David M. Vieth 1973. Pp. 94. (eds.), The Gyldenstolpe Manuscript Miscel- XXIX. Conrad Lindberg, The Earlier Ver- lany of Poems by John Wilmot, Earl of sion of the Wycliffite Bible. Volume 6: Rochester, and other Restoration Authors. Baruch 3.20–end of OT, edited from MS Stockholm, 1967. Pp. xxvii+385. Christ Church 145. Stockholm, 1973. Pp. 393. XVIII. Bertil Johansson, Law and Lawyers XXX. Stieg Hargevik, The Disputed As- in Elizabethan England as Evidenced in the signment of Memoirs of an English Officer Plays of Ben Jonson and Thomas Middleton. to Daniel Defoe. Part I. Stockholm, 1974. Stockholm, 1967. Pp. 65. Pp. 110+lxxiii. XIX. Stanley Gerson, Sound and Symbol in XXXI. Stieg Hargevik, The Disputed As- the Dialogue of the Works of Charles Dick- signment of Memoirs of an English Officer ens. Stockholm, 1967. Pp. xxxii+382. to Daniel Defoe. Part II. Stockholm, 1974. XX. Conrad Lindberg, MS. Bodley 959. Pp. 107. Genesis–Baruch 3.20 in the Earlier Version of XXXII. Christer Johansson, Old English the Wycliffite Bible. Volume 5: Ecclesiasticus Place-Names and Field-Names Containing 48.6–Baruch 3.20. Stockholm, 1969. Pp. 345. lēah. Stockholm, 1975. Pp. 170. XXI. Sven L. Fristedt, The Wycliffe Bible. XXXIII. Sven Jacobson, Factors Influencing Part II. The Origin of the First Revision as the Placement of English Adverbs in Rela- presented in De Salutaribus Documentis. tion to Auxiliaries. A Study in Variation. Stockholm, 1969. Pp. lxxiii+ 212. Stockholm, 1975. Pp. 599. XXII. Arne Rudskoger, Plain. A Study in XXXIV. Ingvar Carlsson, The Pastoral Co-text and Context. Stockholm, 1970. Pp. Care. Edited from British Museum MS. 223. Cotton Otho B.ii. Part I (ff. 1–25va/4). XXIII. Sven Jacobson, Studies in English Stockholm, 1975. Pp. 166. Transformational Grammar, Stockholm, XXXV. Karl Gunnar Lindkvist, A Compre- 1971. Pp. 139. hensive Study of Conceptions of Locality in XXIV. Karin Aijmer, Some Aspects of which English Prepositions Occur. Stock- Psychological Predicates in English. Stock- holm, 1976. Pp. 363. holm, 1972. Pp. 131. XXXVI. Nils-Lennart Johannesson, The Eng- XXV. Karl-Gunnar Lindkvist, The Local lish Modal Auxiliaries: A Stratificational Ac- Sense of the Preposition Over, Above, and count. Stockholm, 1976. Pp. 165. Across Studied in Present-Day English. XXXVII. Bror Danielsson, Cynegetica Stockholm, 1972. Pp. 107. Anglica I. William Twiti, The Art of Hunt- XXVI. Bror Danielsson and Arvid Gabri- ing, 1327. Stockholm, 1976. Pp. 116, 5 elson, Alexander Gill’s Logonomia Anglica plates, and 35 pictures. (1619). Part I. Facsimiles of Gill’s Presenta- XXXVIII. Bertil Johansson, “God Bless tion Copy in the Bodleian Library You, My Dear Miss Nightingale”. Letters (4ºG30Art.). List of Transcribed Words. from Emmy Carolina Rappe to Florence Stockholm, 1972. Pp. 241. Nightingale 1867–1870. Edited with an XXVII. Bror Danielsson, Arvid Gabrielson Introduction, Notes, and two Appendices. and R. C. Alston, Alexander Gill’s Logo- Stockholm, 1977. Pp. 57. XXXIX. Gunnel Tottie, Fuzzy Negation in 1568). Part II. De Recta et Emendata Lin- English and Swedish. Stockholm, 1977. Pp. guae Graecae Pronuntiatione. Stockholm, 71. 1978. Pp. 218. XL. Monica Bjurman, The Phonology of LI. Mats Rydén, An Introduction to the Jacques Bellot’s Le Maistre d’Escole Anglois Historical Study of English Syntax. Stock- (1580). Stockholm, 1977. Pp. 166. holm, 1979, Pp. 58. XLI. Bertil Johansson, The Adapter LII. Sven Jacobson (ed.), Papers from the Adapted. A Study of Sir John Vanbrugh’s Scandinavian Symposium on Syntactic Comedy “The Mistake”. Its Predecessors Variation, Stockholm, May 18–19, 1979. and Successors. Stockholm, 1977. Pp. 75. Stockholm, 1980. Pp. 172. XLII. Kerstin Klasson, Developments in the LIII. Karin Aijmer, Evidence and the De- Terminology of Physics and Technology. clarative Sentence, Stockholm, 1980. Pp. Stockholm, 1977. Pp. 202. 150. XLIII. Mats Rydén, Shakespearean Plant LIV. Rune Nohlgren, English Pronunciation Names. Identifications and Interpretations. according to Petter Moberg (1801 ff.) and Stockholm, 1978. Pp. 117. Other Swedes before 1900. Stockholm, XLIV. Sven Jacobson, On the Use, Meaning, 1981. Pp. 183. and Syntax of English Preverbal Adverbs. LV. Sven Jacobson, Preverbal Adverbs and Stockholm, 1978. Pp. 160. Auxiliaries. A Study of Word Order Change. XLV. Torkel Nöjd, Richard Hodges’s The Stockholm, 1981. Pp. 100. English Primrose (1644). A Study of the LVI. Bror Danielsson, Sir Thomas Smith. Strong-Stressed Vowels and Diphthongs Literary and Linguistic Works (1542, 1549, with Some Regard to A Special Help to 1568). Part III. A critical edition of De Recta Orthographie (1643), The Plainest Direc- et Emendata Linguae Anglicae Scriptione, tions (1649), Most Plain Directions for True- Dialogus. Stockholm, 1983. Pp. 185. Writing (1653). Stockholm, 1978. Pp. LVII. Sven Jacobson (ed.), Papers from the xxiv+194. Second Scandinavian Symposium on Syn- XLVI. Mats Rydén and Lennart A. Björk tactic Variation. Stockholm, May 15–16, (eds.), Studies in English Philology, Linguis- 1982. Stockholm, 1983. Pp. 176. tics and Literature. Presented to Alarik LVIII. Brita Fjelkestam-Nilsson, Also and Rynell 7 March 1978. Stockholm, 1978. Pp. Too. A Corpus-Based Study of Their Fre- 187. quency and Use in Modern English. Stock- XLVII. Anne Charlotte Svinhufvud, A Late holm, 1983. Pp. viii+170. Middle English Treatise on Horses edited LIX. Catherine Sandbach-Dahlström, Be from British Library MS. Sloane 2584 ff. Good Sweet Maid. Charlotte Yonge’s Do- 102–117b. Stockholm, 1978. Pp. 281. mestic Fiction: A Study in Dogmatic Pur- XLVIII. † Ingvar Carlsson, The Pastoral pose and Fictional Form. Stockholm, 1984. Care. Edited from British Museum MS. Pp. v+185. Cotton Otho B.ii. Part II (ff. 25va/4–end). LX. Siv Nässlin, The English Tag Question: Completed by Lars-G. Hallander et al. A Study of Sentences Containing Tags of the Stockholm, 1978. Pp. 200. Type isn’t it?, is it? Stockholm, 1984. Pp. XLIX. Karl-Gunnar Lindkvist, At versus On, x+204. In, By: on the Early History of Spatial At and LXI. Mats Rydén, The English Plant Names Certain Primary Ideas Distinguishing At from in The Grete Herball (1526). A Contribution On, In, By. Stockholm, 1978. Pp. 90. to the Historical Study of English Plant- L. Bror Danielsson, Sir Thomas Smith. Name Usage. Stockholm, 1984. Pp. 110. Literary and Linguistic Works (1542, 1549, LXII. Sven Jacobson, British and American Conference for English Studies. Vol. I. Scouting and Guiding Terminology. A Lexo- Stockholm, 1987. Pp. vii+413. Semantic Study. Stockholm, 1985. Pp. 112. LXXIV. Ishrat Lindblad and Magnus Ljung LXIII. Anne M. Bindslev, Mrs. Humphry (eds.), Proceedings from the Third Nordic Ward. A Study in Late-Victorian Feminine Conference for English Studies. Vol. II. Consciousness and Creative Expression. Stockholm, 1987. Pp. vii+390. Stockholm, 1985. Pp. vi+166. LXXV. Lennart A. Björk, Psychological LXIV. Bengt Oreström (ed.), A Corpus of Vision and Social Criticism in the Novels of Shetland English. Stockholm, 1985. Pp. 119. Thomas Hardy. Stockholm, 1987. Pp. 178. LXV. Sven Jacobson (ed.), Papers from the LXXVI. Nina Witoszek, The Theatre of Third Scandinavian Symposium on Syntactic Recollection. A Cultural Study of the Mod- Variation, Stockholm, May 11–12, 1985. ern Dramatic Tradition in Ireland and Po- Stockholm, 1986. Pp. 180. land. Stockholm, 1988. Pp. 240. LXVI. Catarina Ericsson, A Child Is a LXXVII. Danuta Zadworna-Fjellestad and Child, You Know: the Inversion of Father Lennart A. Björk (eds.), Criticism in the and Daughter in Dickens’s Novels. Stock- Twilight Zone: Postmodern Perspectives on holm, 1986. Pp. vii+94. Literature and Politics. Stockholm, 1990. Pp. LXVII. Pär Hultfors, Reactions to Non- xii+153. Native English. Native English-Speakers’ LXXVIII. Magnus Ljung, A Study of TEFL Assessments of Errors in the Use of English Vocabulary. Stockholm, 1990. Pp. v+425. Made by Non-Native Users of the Language. LXXIX. Pia Norell, Native-Speaker Reac- Part 1. Acceptability and Intelligibility. tions to Swedish Pronunciation Errors in Stockholm, 1986. Pp. xv+246. English: Recognition, Intelligibility and LXVIII. Danuta Zadworna-Fjellestad, Al- Attitude. Stockholm, 1991. Pp. vi+181. ice’s Adventures in Wonderland and Gravi- LXXX. Beatrice Warren, Sense Develop- ty’s Rainbow: A Study in Duplex Fiction. ments. A Contrastive Study of the Develop- Stockholm, 1986. Pp. v+123. ment of Slang Senses and Novel Standard LXIX. Britt Erman, Pragmatic Expressions Senses in English. Stockholm, 1992. Pp. in English. A Study of You know, You see ii+196. and I mean in Face-to-face Conversation. LXXXI. Conrad Lindberg. The Earlier Stockholm, 1987. Pp. ix+238. Version of the Wycliffite Bible. Volume 7: LXX. Mats Rydén and Sverker Brorström, The Gospels, edited from MS Christ Church The Be/Have Variation with Intransitives in 145. Stockholm, 1994. Pp. 269. English. With Special Reference to the Late LXXXII. Cornelia Ilie, What Else Can I Tell Modern Period. Stockholm, 1987. Pp. 265. You? A Pragmatic Study of English Rhetori- LXXI. Pär Hultfors, Reactions to Non- cal Questions as Discursive and Argumenta- Native English. Native English-Speakers’ tive Acts. Stockholm, 1994. Pp. vii+248. Assessments of Errors in the Use of English LXXXIII. Helena Granlund. The Paradox of Made by Non-Native Users of the Language. Self-Love: Christian Elements in George Part 2. Foreigner Role and Interpretation. Eliot’s Treatment of Egoism. Stockholm, Stockholm, 1987. Pp. xiv+237. 1994. Pp. vi+186. LXXII. Alarik Rynell, Antedatings and LXXXIV. Gunnel Melchers and Nils- Additions for OED from the Catalogue of Lennart Johannesson (eds.), Nonstandard Prints of Political and Personal Satire in the Varieties of Language. Papers from the British Museum. Stockholm, 1987. Pp. 184. Stockholm Symposium 11–13 April, 1991. LXXIII. Ishrat Lindblad and Magnus Ljung Stockholm, 1994. Pp. vi+220. (eds.), Proceedings from the Third Nordic LXXXV. Gunnel Melchers and Beatrice XCVI. Annika Ljung-Baruth, A Steady Warren (eds.), Studies in Anglistics. Stock- Flameless Light. The Phenomenology of holm, 1995. Pp. vii+316. Realness in Dorothy Canfield Fisher’s The LXXXVI. Yvonne Martinsson, Eroticism, Brimming Cup, Her Son’s Wife and Rough- Ethics and Reading: Angela Carter in Dia- Hewn. Stockholm, 2002. Pp. vi + 169. logue with Roland Barthes. Stockholm, XCVII. Joakim Sigvardson, Immanence and 1996. Pp. v+140. Transcendence in Thomas Pynchon’s Mason LXXXVII. Conrad Lindberg, The Earlier & Dixon. A Phenomenological Study. Version of the Wycliffite Bible. Volume 8: Stockholm, 2002. Pp. x+158. The Epistles etc., edited from MS Christ XCVIII. Conrad Lindberg, King Henry’s Church 145. Stockholm, 1997. Pp. 376. Bible. MS Bodley 277. The Revised Version LXXXVIII. Gabriella Rundblad. Shallow of the Wyclif Bible. Volume III: Proverbs–II Brooks and Rivers Wide. A Study of Lexical Maccabees. Stockholm, 2002. Pp. 550. and Semantic Change in English Nouns De- XCIX. Anka Ryall and Catherine Sandbach- noting ‘Watercourse’. Stockholm, 1998. Pp. Dahlström (eds), Mary Wollstonecraft’s xiv+212. Journey to Scandinavia: Essays. Stockholm, LXXXIX. Conrad Lindberg, King Henry’s 2003. Pp. v+248. Bible. MS Bodley 277. The Revised Version C. Conrad Lindberg, King Henry’s Bible. of the Wyclif Bible. Volume I: Genesis– MS Bodley 277. The Revised Version of the Ruth. Stockholm, 1999. Pp. 406. Wyclif Bible. Volume IV: The New Testa- XC. Dee Drake, Searing Apparent Surfaces: ment. Stockholm, 2004. Pp. 361. Infernal Females in Four Early Works of CI. Harald Fawkner. Grasses that Have No William Blake. Stockholm, 2000. Pp. 178. Fields. From Gerald Murnane’s Inland to a XCI. Marion Helfer Wajngot, The Birthright Phenomenology of Isogonic Constitution. and the Blessing: Narrative as Exegesis in Stockholm, 2006. Pp. 108. Three of Thackeray’s Later Novels. Stock- CII. Conrad Lindberg. A Manual of the holm, 2000. Pp. 211. Wyclif Bible, Including the Psalms. Dedi- XCII. Magnus Ljung (ed.), Language Struc- cated to the Memory of Sven L. Fristedt. ture and Variation. Stockholm, 2001. Pp. Stockholm, 2007. Pp. 224. v+212. CIII. Nils-Lennart Johannesson & David C. XCIII. Inger Björkblom, The Plane of Un- Minugh (eds), Selected Papers from the 2006 createdness: a Phenomenological Study of and 2007 Stockholm Metaphor Festivals. Anita Brookner’s Late Fiction. Stockholm, Stockholm, 2008. Pp. 170. 2001. Pp. viii+167. Six plates. CIV. Nils-Lennart Johannesson, Gunnel XCIV. Conrad Lindberg, King Henry’s Melchers & Beyza Björkman (eds), Of But- Bible. MS Bodley 277. The Revised Version terflies and Birds, of Dialects and Genres. of the Wyclif Bible. Volume II: I Kings– Essays in honour of Philip Shaw. Stock- Psalms. Stockholm, 2001. Pp. 570. holm, 2013. Pp. 399. XCV. Harriet Sharp, English in Spoken CV. Nils-Lennart Johannesson & David C. Swedish: A Corpus Study of Two Discourse Minugh (eds.), Selected Papers from the Domains. Stockholm, 2001. Pp. 212. 2008 Stockholm Metaphor Festival. Stock- holm, 2013. Pp. 221. ACTA UNIVERSITATIS STOCKHOLMIENSIS Corpus Troporum Romanica Stockholmiensia Stockholm Cinema Studies Stockholm Contributions in Geology Stockholm Fashion Studies Stockholm Oriental Studies Stockholm Slavic Studies Stockholm Studies in Baltic Languages Stockholm Studies in Comparative Religion Stockholm Studies in Economic History Stockholm Studies in English Stockholm Studies in Ethnology Stockholm Studies in Film History Stockholm Studies in History Stockholm Studies in History of Literature Stockholm Studies in Human Geography Stockholm Studies in Linguistics Stockholm Studies in Modern Philology. N.S. Stockholm Studies in Musicology Stockholm Studies in Philosophy Stockholm Studies in Russian Literature Stockholm Studies in Scandinavian Philology. N.S. Stockholm Studies in Sociology. N.S. Stockholm University Demography Unit – Dissertation Series Stockholmer Germanistische Forschungen Studia Fennica Stockholmiensia Studia Juridica Stockholmiensia Studia Latina Stockholmiensia