schema, for example, can be understood as the superimposition of a path whose goal and source are identical points in space due to iteration. This ‘‘Gestalt’’ grouping of image schemas with other image schemas leads to the tentative con- clusion that some image schemas are perceptually more primary (e.g., path), while others suggest a more complex structure; thus, distinguishing between recurring perceptual experience and gestalt complexes of perceptual experiences is crucial, despite their treatment as synonyms in much of the literature. Distinguishing the developmental trajectory of image schemas may be another way of grouping them. Some image schemas may be developmentally more basic than others. According to Mandler (1992), conceptual development arises from perceptual analysis; by this mechanism, perceptual information is conceptualized, with the resulting notions of animacy, inanimacy, agency, and containment guiding the initial phases of conceptual development. Image schemas for path, compulsion, link, counterforce, contact, surface, and object may be onto- genetically basic, because each requires one or more of these forms of perceptual analysis. Now consider what known image schemas have in common. All image schemas can be construed as dynamic or static scenes, as processes or states (see Cienki 1997, for a review). For example, we can construe the experience of balance as a state of equilibrium or as an act of maintaining balance. The static versus dynamic characteristics of image schemas references Langacker’s (1987: 145) distinction between summary and sequential scanning. When we construe a com- plex scene as one in which all facets are ‘‘conceived coexistent and simultaneously available,’’ we are relying on static realizations of image schemas; then again, when we construe a complex scene as a series of states in which one successively trans- forms into another, we are relying on dynamic realizations of image schemas. Krzeszowski (1993: 310) discusses in some detail an important characteristic shared by all image schemas: the plus-minus parameter. That is, all image schemas exhibit a bipolar property of conferring positive or negative associations. In other words, image schemas all have euphoric or dysphoric characteristics, and it is this ‘‘axiological parameter’’ that is ‘‘responsible for the dynamism of the metaphor- ization processes inherent in the formation of concepts based on the relevant schemata’’ (310). Thus, balance is positive/plus, while imbalance is negative/minus; whole is positive while part is negative. These euphoric and dysphoric properties are imperative for forming axiological concepts like ‘good’, ‘beautiful’, ‘true’, ‘bad’, ‘ugly’, and ‘false’ (325). Even with these distinguishing and common features, image schemas seem to exhibit such a wide variety of instantiations that systematic investigation may seem impossible. This is where Kreitzer’s (1997) study of schematization may come in handy. Taking the case of over as his point of departure, Kreitzer distinguishes between three levels of schematization: the component level, the relational level, and the integrative level. The component level refers to particular schematized elements of a spatial scene, such as surfaces, lines, and points. They are geometrically schematic, 230 todd oakley allowing for an image schema to apply to a wide variety of objects and relations. For example, the support and contact schemas, realized linguistically by the prepo- sition on, are schematized as points of a mass in contact with a surface. The rela- tional level organizes components into specific spatial relations. The prototype for over requires motion and a trajectory traversing the boundaries of a landmark. This means that contact with the landmark is not relevant at the component or relational level of schematization. At the integrative level, multiple image schemas unite. Kreitzer gives the following example, taken from Lakoff (1987): The city clouded over. In this case, the path schema integrates with a static covering schema, but the crucial point is that the covering schema is not a part of the path schema at the relational level, for the construed relation is static not dynamic, hence no end-point focus transformation occurs (Kreitzer 1997: 307). The main advantage of Kreitzer’s levels is that it offers principled criteria for specifying different components of image schemas, with different lexical meanings arising from different integrations of re- lational schemas. For instance, over exhibits three different relational schemas: the first specifying a static relation along the vertical axis; a second specifying a dynamic relation along the vertical and horizontal axes; and a third specifying a static relation defined egocentrically. Kreitzer argues that the component level determines the basic primitives of spatial conceptualization. The term ‘‘spatial conceptualization’’ brings up an interesting issue regarding how image schemas are understood outside the cognitive linguistic community. Jackendoff’s (1996) treatment of the ‘‘linguistic-spatial interface’’ makes mention of image schemas as ‘‘abstract representations from which a variety of images can be generated’’ (9). In his scheme of representational modularity, Jackendoff places image schemas within the module of spatial representation, which is largely ‘‘in- visible’’ to conceptual structure. Strictly speaking, image schemas are useful de- scriptions of spatial representation, providing three-dimensional models of spatial properties, but their influence on conceptual structure and linguistic structure are minimal. Jackendoff’s theory of representation modularity is at once an acknowl- edgment of the power of image schemas as a theoretical notion and a denial that image schemas play a significant role in conceptualization and grammar. In addition, Jackendoff (1996: 26) is quick to argue that ‘‘it is not obvious that places and paths are encoded in imagistic representation because we do not literally see them except when dotted lines are drawn in cartoons.’’ This seemingly off- handed remark, however, brings up an important issue regarding the graphic rep- resentation of image schemas, namely that these conventions may subtly channel cognitive linguistic investigations in ways that can bias theory and analysis in certain directions. This argument is made explicitly in Mandler (1992) and implicitly in Dewell (1994). Diagrams for motion in notation systems for Cognitive Grammar usually in- clude straight lines. But such a representation does not help us distinguish between animate motion, self-propelled motion, and inanimate and caused motion, distinc- tions that are essential for perceptual analysis, as outlined in Mandler (1992). The manner in which something traverses a path is an important, if unstated, element of image schemas 231 image schema theory; that is, there are probably distinct schemas for animate and inanimate motion. Dewell (1994) argues that typical accounts of over in Cognitive Linguistics posit several features, such as the shape of and contact with a landmark, as well as pos- iting across and above as subschemas, which Dewell regards as separate schemas altogether. Although Dewell does not make this connection explicit, one can see that Lakoff ’s and Brugman’s decision to make ‘flat trajectory’ examples of over the prototype stem, in part, from the tendency to represent trajectories as straight lines. But, as Dewell points out, the best linguistic evidence (e.g., The dog jumped over the fence) suggests that the most typical examples of over involve an arched path schema, and from it he accounts for all the uses of over without specifying extra- neous features, such as ‘contact’. Across and above are not subschemas of over but distinct schemas with their own inference generating capacities and grammatical realizations. Close examination of Lakoff’s (1987: 419) case study reveals that he builds his analysis around an initial ‘above and across’ representation, with an oval marked trajector placed along a straight dotted arrow which runs from left to right and is placed over a square marked landmark, and that all subsequent graphic rep- resentations and analyses issue from that basic representation. Dewell, on the other hand, builds his analysis around a semicircular path with an arched arrow partially enclosing a rectangle (1994: 355). To summarize, graphic conventions of many image schema–based studies represent motion in a straight line as the default pro- totype, but as Mandler and Dewell argue, motion and paths are not prototypically straight. It is open for debate as to how the conventions of image-schematic rep- resentations actually bias theory and analysis. These two studies at least hint that this is an issue, as are the others mentioned above, that is worth our attention. REFERENCES Bailey, David, Jerome Feldman, Srini Narayanan, and George Lakoff. 1997. Modeling embodied lexical development: Neural theory of language publications. http://www .icsi.berkeley.edu/NTL/papers/cogsci97.pdf. Accessed February 8, 2001. Boers, Frank, and Murielle Demecheleer. 1997. A few metaphorical models in (Western) economic discourse. In Wolf-Andreas Liebert, Gisela Redeker, and Linda Waugh, eds., Discourse and perspective in cognitive linguistics 115–29. Amsterdam: John Benjamins. Casad, Eugene H. 1998. Lots of ways to GIVE in Cora. In John Newman, ed., The linguistics of giving 135–74. Amsterdam: John Benjamins. Cienki, Alan. 1997. Some properties and groupings of image schemas. In Marjolijn Ver- spoor, Kee Dong Lee, and Eve Sweetser, eds., Lexical and syntactical constructions and the construction of meaning 3–15. Amsterdam: John Benjamins. Cienki, Alan. 1998. straight: An image schema and its metaphorical extensions. Cognitive Linguistics 9: 107–49. D’Andrade, Roy. 1995. The development of cognitive anthropology. Cambridge: Cambridge University Press. 232 todd oakley Delbecque, Nicole. 1995. Towards a cognitive account of the use of the prepositions por and para in Spanish. In Eugene H. Casad, ed., Cognitive linguistics in the Red- woods: The expansion of a new paradigm in linguistics 249–318. Berlin: Mouton de Gruyter. Dewell, Robert B. 1994. Over again: Image schema transformations in semantic analysis. Cognitive Linguistics 5: 351–80. Ekberg, Lena. 1995. The mental manipulation of the vertical axis: How to go from ‘‘up’’ to ‘‘out’’ or from ‘‘above’’ to ‘‘behind.’’ In Marjolijn Verspoor, Kee Dong Lee, and Eve Sweetser, eds., Lexical and syntactical constructions and the construction of meaning 69–88. Amsterdam: John Benjamins. Freeman, Margaret H. 1995. Metaphor making meaning: Dickinson’s conceptual universe. Journal of Pragmatics 24: 643–66. Freeman, Margaret H. 2002. Momentary stays, exploding forces: A cognitive linguistic approach to the poetics of Emily Dickinson and Robert Frost. Journal of English Linguistics 30: 73–90. Gibbs, Raymond W., Jr. 1994. The poetics of mind: Figurative thought, language, and under- standing. Cambridge: Cambridge University Press. Gibbs, Raymond W., Jr, Dinara Beitel, Michael Harrington, and Paul Sanders. 1994. Taking a stand on the meaning of stand: Bodily experience as motivation for polysemy. Journal of Semantics 11: 231–51. Gibbs, Raymond W., Jr., and Herbert L. Colston. 1995. The cognitive psychological reality of image schemas and their transformations. Cognitive Linguistics 6: 347–78. Jackendoff, Ray. 1996. The architecture of the linguistic-spatial interface. In Paul Bloom, Mary A. Peterson, Lynn Nadel, and Merrill F. Garrett, eds., Language and space 1–30. Cambridge, MA: MIT Press. Johnson, Mark. 1987. The body in the mind: The bodily basis of meaning, imagination, and reason. Chicago: University of Chicago Press. Johnson, Mark. 1993. Moral imagination: Implications of cognitive science for ethics. Chicago: University of Chicago Press. Kant, Immanuel. [1781] 1990. Critique of pure reason. Translated by J. M. D. Meiklejohn. Buffalo, NY: Prometheus Books. Kemmer, Suzanne, and Michael Barlow. 2000. Introduction. In Suzanne Kemmer and Michael Barlow, eds., Usage-based models of language vii–xxviii. Stanford, CA: CSLI Publications. Kreitzer, Anatol. 1997. Multiple levels of schematization: A study in the conceptualization of space. Cognitive Linguistics 8: 291–325. Krzeszowski, Tomasz P. 1993. The axiological parameter in preconceptual image schemata. In Richard Geiger and Brygida Rudzka-Ostyn, eds., Conceptualization and mental processing in language 307–29. Berlin: Mouton de Gruyter. Lakoff, George. 1987. Women, fire, and dangerous things: What categories reveal about the mind. Chicago: University of Chicago Press. Lakoff, George, and Mark Johnson. 1980. Metaphors we live by. Chicago: University of Chicago Press. Lakoff, George, and Rafael Nu ´ n ˜ ez. 2000. Where mathematics comes from: How the embodied mind brings mathematics into being. New York: Basic Books. Lakoff, George, and Mark Turner. 1989. More than cool reason: A field guide to poetic metaphor. Chicago: University of Chicago Press. Langacker, Ronald W. 1987. Foundations of cognitive grammar. Vol. 1, Theoretical pre- requisites. Stanford, CA: Stanford University Press. image schemas 233 Langacker, Ronald W. 1991. Foundations of cognitive grammar. Vol. 2, Descriptive appli- cation. Stanford, CA: Stanford University Press. Mandler, Jean. 1992. How to build a baby: II. Conceptual primitives. Psychological Review 99: 597–604. Matsumoto, Yo. 1995. From attribution/purpose to cause: Image schema and grammati- calization of some cause markers in Japanese. In Marjolijn Verspoor, Kee Dong Lee, and Eve Sweetser, eds., Lexical and syntactical constructions and the construction of meaning 287–307. Amsterdam: John Benjamins. Narayanan, Srini. 1999. Moving right along: A computational model of metaphoric reasoning about events. Proceedings of the National Conference on Artificial Intelli- gence (AAAI-99) 121–28. Menlo Park, CA: AAAI Press. Palmer, Gary B. 1996. Toward a theory of cultural linguistics. Austin: University of Texas Press. Pauwels, Paul. 1995. Levels of metaphorization: The case of put. In Louis Goossens, ed., By word of mouth: Metaphor, metonymy and linguistic action in a cognitive perspective 125–58. Amsterdam: John Benjamins. Regier, Terry. 1996. A model of the human capacity for categorizing spatial relations. Cognitive Linguistics 6: 63–88. Rhee, Seongha. 2002. Semantic changes of English prepositions against a grammaticali- zation perspective. Language Research 38: 563–83. Romaine, Suzanne. 1996. War and peace in the global greenhouse: Metaphors we live and die by. Metaphor and Symbolic Activity 1996: 175–94. Serra-Borneto, Carlo. 1995a. Two-way prepositions in German: Image and constraints. In Marjolijn Verspoor, Kee Dong Lee, and Eve Sweetser, eds., Lexical and syntactical constructions and the construction of meaning 187–204. Amsterdam: John Benjamins. Serra-Borneto, Carlo. 1995b. Liegen and stehen in German: A study of horizontality and verticality. In Eugene H. Casad, ed., Cognitive linguistics in the Redwoods: The ex- pansion of a new paradigm in linguistics 459–505. Berlin: Mouton de Gruyter. Sinha, Chris, and Kristine Jensen de Lo ´ pez. 2000. Language, culture, and the embodiment of spatial cognition. Cognitive Linguistics 11: 17–41. Smith, Michael B. 1999 . From instrument irrealis: Motivating some grammaticalized senses of the Russian instrumental. In Katarzyna Dziwirek, Herbert Coats, and Cynthia M. Vakareliyska, eds., Annual workshop on formal approaches to Slavic linguistics. Vol. 7, The Seattle meeting 1998 413–33. Ann Arbor: Michigan Slavic Publications. Smith, Michael B. 2002. The polysemy of German es: Iconicity and the notion of con- ceptual distance. Cognitive Linguistics 13: 67–112. Talmy, Leonard. 1983. How language structures space. In Herbert L. Pick, Jr., and Linda P. Acredolo, eds., Spatial orientation: Theory, research, and application 225–82. New York: Plenum Press. Turner, Mark. 1987. Death is the mother of beauty: Mind, metaphor, criticism. Chicago: University of Chicago Press. Turner, Mark. 1991. Reading minds: The study of English in the age of cognitive science. Princeton, NJ: Princeton University Press. Turner, Mark. 1992. Language is a virus. Poetics Today 13: 725–36. Verspoor, Marjolijn. 1995. Predicate adjuncts and subjectification. In Marjolijn Verspoor, Kee Dong, and Eve Sweetser, eds., Lexical and syntactical constructions and the con- struction of meaning: Proceedings of the bi-annual ICLA meeting in Albuquerque, July 1995 433–49. Amsterdam: John Benjamins. 234 todd oakley Wagner, Susan, Ellen Winner, Diane Cicchetti, and Howard Gardner. 1981. Metaphorical mappings in human infants. Child Development 52: 728–31. Watters, James K. 1995. Frames and the semantics of applicatives in Tepehua. In Eugene H. Casad, ed., Cognitive linguistics in the Redwoods: The expansion of a new paradigm in linguistics 971–96. Berlin: Mouton de Gruyter. Williams, Darrell. 1992. English comparative compounds with OVER, UNDER and OUT. Proceedings of the Eastern States Conference on Linguistics 9: 272–81. image schemas 235 chapter 10 METONYMY klaus-uwe panther and linda l. thornburg 1. Introduction Metonymy is a cognitive phenomenon—not just a figure of speech—whose role in the organization of meaning (semantics), utterance production and interpretation (pragmatics), and even grammatical structure is considerable. The same meto- nymic principles that relate different senses of a word serve to create and retrieve novel meanings in actual language use. The ubiquity of metonymy can be inter- preted as an indication that there is a continuum between linguistic meaning and communicative use rather than a strict division of labor between two autonomous components, semantics and pragmatics. Furthermore, the interpretation of gram- matical structure (construction meaning) seems to be sensitive to metonymic principles. Finally, metonymic processes play a crucial role in semantic change and in grammaticalization. This chapter is organized as follows: after a brief—and necessarily nonexhaustive—summary of the rhetorical tradition in section 2, various cognitive linguistic approaches to metonymy are discussed in section 3. A working definition of metonymy is developed in section 4 , which is applied in the subsequent sections. Section 5 reports some work that demonstrates the interaction of metonymy with metaphor and the experiential grounding of metonymy. Section 6 is concerned with the role of metonymy in referential, predicational, propositional, and illo- cutionary acts. Section 7 considers metonymy in relation to pragmatic inferenc- ing, that is, implicature and explicature, and discusses some of its discourse- pragmatic functions. In section 8, the interaction of metonymy with grammatical structure is explored. Section 9 reports on work that compares the exploitation of metonymies cross-linguistically. Section 10 describes the role of metonymy in dia- chronic change, in particular, semantic change and grammaticalization. Section 11 briefly considers the role of metonymy in language production, comprehension, and acquisition. Section 12 concludes the chapter with a discussion of unresolved problems, an analysis of the taxonomic structure of one high-level metonymy, and suggestions for future research. 2. The Rhetorical Tradition Metonymy (Greek metonumipa, Latin denominatio) is one of the major figures of speech recognized in classical rhetoric. One of the earliest definitions of meton- ymy is attributed to the treatise Rhetorica ad Herennium (see Koch 1999: 140). The anonymous author characterizes metonymy as ‘‘a trope that takes its expression from near and close things [‘ab rebus propinquis et finitimis’] by which we can comprehend a word that is not denominated by its proper word’’ (translation by Koch 1999: 141). This ancient characterization already points to the notions of con- tiguity and association that have ever since been criterial in distinguishing meton- ymy from metaphor. There is a rich tradition of research on metonymy in the historical-philological tradition of linguistics. As pointed out by Geeraerts (1988), the psychological ori- entation of much of nineteenth-century philology, such as the works of Michel Bre ´ al and Hermann Paul, is theoretically very close to present-day Cognitive Semantics. Furthermore, the study of etymology almost inevitably leads to an interest in the general principles of semantic change including the role of metonymy (and met- aphor) in the development of new meanings. The concept of metonymy has remained remarkably constant since antiquity: a typical twentieth-century definition of metonymy that is not essentially different from the one given by the author of Rhetorica ad Herennium is found in Geeraerts (1994: 2477): ‘‘[Metonymy is] a semantic link between two senses of a lexical item that is based on a relationship of contiguity between the referents of the expression in each of those senses.’’ Traditionally, then, metonymy has been regarded as a stand for relation in which the name of one thing (henceforth, the source or vehicle) is used to refer to another thing (henceforth, the target) with which it is associated or to which it is contiguous. This view can be called the substitution theory of metonymy. A corol- lary of the substitution theory is that the source and the target are, at some level of analysis, considered to be equivalent ways of picking out the same referent. For example, in the sentence Buckingham Palace issued a statement this morning, the place name Buckingham Palace (source) may be said to stand for the British queen or one of her spokespersons (target). Under this view, the source expression metonymy 237 indirectly achieves the same referential purpose as the more direct referring ex- pression the Queen. The substitution theory is, however, too simplistic in at least two respects. First, it typically focuses only on cases of referential metonymy, neglecting evidence that metonymy is also found on the predicational and illo- cutionary levels (see section 6). Second, as Radden and Ko ¨ vecses ( 1999: 18) point out, metonymy involves more than just an operation of substitution. For example, in She is just a pretty face, the noun phrase a pretty face is not used referentially but predicatively; as well, it is not just a substitute expression for a pretty person but also highlights the prettiness of the person’s face, from which the prettiness of the person can be inferred. Thus, the above sentence expresses more content than ‘She is just a pretty person’. The attempt to develop a sufficiently narrow definition of metonymy leads to the question of how it is to be delimited from other figures of speech and thought. The two tropes in relation to which metonymy is normally seen are metaphor and synecdoche. In traditional rhetoric, synecdoche is regarded as ‘‘quite distinct from metonymy’’ (Bredin 1984: 46). More recently, however (see, e.g., Jakobson [1956] 1971), synecdoche has come to be considered a subtype of metonymy. Jakobson’s by now famous distinction between metaphor and metonymy links the former to paradigmatic selection in terms of similarity and contrast and the latter to the syntagmatic combination of semantically contiguous elements. A rather idiosyn- cratic theory has been put forward by the Groupe de Lie ` ge or Groupe m (Dubois et al. 1970), where synecdoche is considered as the fundamental trope with both metaphor and metonymy as derivative categories (Schofer and Rice 1977; Bredin 1984: 45). 3. Metonymy in Cognitive Linguistics In Cognitive Linguistics, metonymy and synecdoche are often believed to instan- tiate the same conceptual phenomenon (Lakoff 1987). 1 Lakoff and Johnson (1980) see metonymy (including synecdoche) as a predominantly referential shift phe- nomenon within one cognitive domain—much in line with the traditional concep- tion of metonymy (see section 2 above). In contrast, they regard metaphor as a mapping from one conceptual domain onto another distinct conceptual domain, where the structure of the target is isomorphic to that of the source (Invariance Hypothesis). Most contemporary accounts in Cognitive Linguistics have built on Lakoff and Johnson’s (1980) original distinction between metaphor as a cross-domain mapping and of metonymy as a mapping within one cognitive domain. One influential at- tempt to elaborate Lakoff and Johnson’s characterization is Croft’s (1993: 348) pro- posal that metaphor is ‘‘a mapping between two domains that are not part of the 238 klaus-uwe panther and linda l. thornburg same matrix,’’ whereas metonymy is a mapping within one ‘‘domain matrix.’’ The notion of domain matrix goes back to Langacker’s (1987) insight that the meaning of an expression can often only be determined against the background of a set of over- lapping domains that jointly serve as a base against which the meaning of an expres- sion is profiled. Croft (1993: 348) defines metonymy as a process of domain high- lighting ‘‘since the metonymy makes primary a domain that is secondary in the literal meaning.’’ Thus, in the utterance The Times hasn’t arrived yet, t he noun phr ase The Times metonymically highlights a subdomain of the semantic frame it evokes—such as, a journalist writing for the newspaper—which is usually only secondary. This case is contrasted with the interpretation of the definite description this book in This book is heavy, where both subdomains of book as a physical object and as a bearer of content are argued to be of equal importance and therefore nonmetonymic. Ruiz de Mendoza (2000) proposes that metonymic mappings, which are usually considered to be whole-part, part-whole, or part-part mappings, can be reduced to two kinds: either the source of the metonymic operation is in the target (‘‘source- in-target’’ metonymy) or the target is in the source (‘‘target-in-source’’ metonymy). For example, for The ham sandwich is waiting for his check, Ruiz de Mendoza argues (2000: 114–15) that the contiguity link between ham sandwich and restaurant customer is not a part-part relation in the domain restaurant but rather a source-in-target metonymy where the ham sandwich is conceptualized as being within the target domain the customer. As an example of target-in-source me- tonymy, Ruiz de Mendoza (2000: 127) cites I broke the window, which in most sit- uations conveys that it is not the window as a whole but typically only the win- dowpane that was broken. Various scholars have claimed that metonymy is as pervasive a phenomenon in language and thought as metaphor. Lakoff (1987) stresses the cognitive importance of metonymic models alongside propositional, image-schematic, and metaphorical cognitive models (called ‘‘Idealized Cognitive Models,’’ or ICMs), and in recent vol- umes (e.g., Panther and Radden 1999b; Barcelona 2000a; Dirven and Po ¨ rings 2002), it has been argued that the conceptual and linguistic significance of metonymy is comparable to that of metaphor. Furthermore, some scholars (e.g., Barcelona 2000b; Radden 2000; Ruiz de Mendoza 2000; Ruiz de Mendoza and Campo 2002) have claimed that the borderline between metaphor and metonymy is blurred. Never- theless, there are clear and agreed-upon cases of metonymy (and metaphor) and it is on these prototypical cases that the present chapter focuses. A widely accepted definition of metonymy inspired by Langacker (1993)isthe one proposed by Radden and Ko ¨ vecses (1999: 21): ‘‘Metonymy is a cognitive process in which one conceptual entity, the vehicle, provides mental access to another con- ceptual entity, the target, within the same cognitive model.’’ The notion of cognitive model is taken in its broadest sense, encompassing three ‘‘ontological realms’’ (23): concepts, forms (especially linguistic), and things and events in the ‘‘real world.’’ Over these realms, five potential metonymic relations are defined: (i) the sign relation between form and concept (e.g., the relation between the form house and the concept house); (ii) three ‘‘referential’’ relations (Form-Thing/Event, Concept-Thing/Event, metonymy 239 . cognitive account of the use of the prepositions por and para in Spanish. In Eugene H. Casad, ed., Cognitive linguistics in the Red- woods: The expansion of a new paradigm in linguistics 249–318 Mark. 1987. Death is the mother of beauty: Mind, metaphor, criticism. Chicago: University of Chicago Press. Turner, Mark. 1991. Reading minds: The study of English in the age of cognitive science. Princeton,. corol- lary of the substitution theory is that the source and the target are, at some level of analysis, considered to be equivalent ways of picking out the same referent. For example, in the sentence