Showing papers in "Cognitive Linguistics in 1995"

The cognitive psychological reality of image schemas and their transformations

Abstract: One of the important theoretical ideas in cognitive semantics is t hat image Schemas and their transformations provide pari of the foundation for ihought, reasoning, and Imagination. Image Schemas are different patterns of recurring bodily experiences that emerge throughout sensorimotor activity andfrom our perceptual understanding ofactions andevents in the world. Our aim in this paper is to discuss some of the empirical evidence from psycholinguistics, cognitive psychology, and developmental psychology that is consistent with the idea that image Schemas and their transformations play important roles in human cognition. This experimental research was not conducted and has not generally been considered in terms ofthe cognitive linguistic ideas on image Schemas. However, a large body of research can be interpreted äs supporting the claim that image Schemas are indeed psychologically real andfunction in many aspects of how people process linguistic and nonlinguistic Information. Our review suggests possible ways of integrating this research with thefindings on linguistic structure and meaning in cognitive semantics. One of the important Claims of cognitive semantics is that much of our knowledge is not static, propositional and sentential, but is grounded in and structured by various patterns of our perceptual interactions, bodily actions, and manipulations of objects (Johnson 1987, 1993; Lakoff 1987, 1990; Talmy 1988). These patterns are experiential gestalts, called image Schemas, that emerge throughout sensorimotor activity äs we manipulate objects, Orient ourselves spatially and temporally, and direct our perceptual focus for various purposes (Johnson 1991). Studies in cognitive linguistics suggest that over two dozen different image Schemas and several image Schema transformations appear regularly in people's everyday thinking, reasoning, and Imagination (Johnson Cognitive Linguistics 6-4 (1995), 347-378 0936-5907/95/0006-0347 © Walter de Gruyter 348 R. W. Gibbs, Jr. and H. L. Colston 1987; Lakoff 1987). Among these are the schematic structures of CONTAINER, BALANCE, SOURCE-PATH-GOAL, PATH, CYCLE, ATTRACTION, CENTER-PERIPHERY, and LINK. These image Schemas cover a wide ränge of experiential structures that are pervasive in experience, have internal structure, and can be metaphorically elaborated to provide for our understanding of more abstract domains. For example, cognitive linguistic research has examined how image Schemas are used to create grammatical forms (Langacker 1987, 1991), to represent the underlying meaning that relates the seemingly disparate senses of prepositions (Brugman and Lakoff 1988; Vandeloise 1993), to motivate verb-particle constructions, such äs those focusing on up and out (Lindner 1983), adverbs, such äs very (Brugman 1984), certain verbs, such äs take (Norvig and Lakoff 1987), äs well äs to explain the many kinds of cognitive relationships that can form the basis of the extension of a category such äs Japanese hon (Lakoff 1987). More recent investigations from linguistics and philosophy examined the role that image Schemas have in motivating abstract metaphorical concepts, such äs causation, death, and morality (Johnson 1993; Lakoff 1990; Lakoff and Turner 1989; Turner 1991). Although these studies provide important evidence on image Schemas in everyday thought and linguistic understanding, the question remains äs to whether there exists independent empirical evidence on the psychological reality of image Schemas. Our aim in this paper is to describe some of the Undings from psycholinguistics, cognitive psychology, and developmental psychology that, in our view, support the claims of cognitive semantics about image Schemas and their transformations. There are two important reasons for considering this psychological evidence. First, cognitive linguists, following the cognitive commitment to construct theories that are consistent with what is known about the mind and brain (Lakoff 1990, 1993), should be aware of the experimental fmdings from neighboring disciplines, especially data that bear on the possible connections between perception, thought, and language. Second, psychologists are sometimes skeptical about theoretical notions from linguistics that are primarily based on an individual analyst's intuitions about linguistic structure and behavior. One of the main reasons for conducting experiments with large groups of people is to minimize the uncertainty in making inferences about thought and behavior in whole populations of people. We do not entirely agree with the skepticism of psychologists about the theoretical Claims of cognitive linguists (e.g., Kennedy and Vervaeke 1993). Yet we think there exist different kinds of empirical evidence from psychology that both psychologists and cognitive linguists should be aware of regarding the importance of image Schemas in ordinary cognitive Psychology of Image Schemas 349 functioning. This paper describes some of this evidence. We begin by first elaborating the notion of image Schemas and how they are transformed. We then review work from psycholinguistics that has explicitly examined how image Schemas motivate people's understanding of word meaning. The next section of the paper describes work from cognitive psychology that seems quite consistent with Claims for the importance of image Schemas in everyday cognition. We then review work from developmental psychology that also supports the cognitive reality of image Schemas. The final section discusses the significance of the different work from psychology for future studies in cognitive linguistics. Image Schemas and their transformations Image Schemas can generally be defined äs dynamic analog representations of spatial relations and movements in space. Even though image Schemas are derived from perceptual and motor processes, they are not themselves sensorimotor processes. Instead, image Schemas are \"primary means by which we construct or constitute order and are not mere passive receptacles into which experience is poured\" (Johnson 1987: 30). In this way, image Schemas are different from the notion of Schemata traditionally used in cognitive science, which are abstract conceptual and propositional event structures (see Rumelhart 1980). By contrast, image Schemas are imaginative and nonpropositional in nature and operate äs organizing structures of experience at the level of bodily perception and movement. Image Schemas exist across all perceptual modalities, something that must hold for there to be any sensorimotor coordination in our experience. As such, image Schemas are at once visual, auditory, kinesthetic, and tactile. We can illustrate what is meant by the notion of image Schema, and how its internal structure is projected onto new domain via metaphor, by considering the BALANCE Schema (Johnson 1987). The idea of balance is something that is learned \"with our bodies and not by grasping a set of rules\" (Johnson, 1987: 74). Balancing is such a pervasive part of our bodily experience that we are seldom aware of its presence in everyday life. We come to know the meaning of balance through the closely related experiences of bodily equilibrium or loss of equilibrium. For example, a baby Stands, wobbles, and drops to the floor. It tries again and again, äs it learns how to maintain a balanced erect posture. A young boy struggles to stay up on a two-wheeled bicycle äs he learns to keep bis balance while riding down the street. Bach of us has experienced occasions when we have too much acid in our stomachs, when our hands get cold, our heads feel too hot, our bladders feel distended, our sinuses become swollen, and our mouths feel dry. In these and numerous other ways we 350 K W. Gibbs, Jr. and H. L. Colston learn the meanings of lack of balance or equilibrium. We respond to imbalance and disequilibrium by warming our hands, giving moisture to our mouths, draining our bladders, and so forth until we feel balanced once again. Our BALANCE image Schema emerges, then, through our experiences of bodily equilibriums and disequilibriums and of maintaining our bodily Systems and functions in states of equilibrium. We refer to these recurring bodily experiences äs image Schemas to emphasize means of structuring particular experiences schematically so that we can give order and connectedness to our perceptions and conceptions (Johnson 1987). One of the most interesting things about image Schemas is that they motivate important aspects of how we think, reason, and ünagine. The same image Schema can be instantiated in many different kinds of domains because the internal structure of a single Schema can be metaphorically understood. Our BALANCE image Schema, to continue with this example, is metaphorically elaborated in a large number of abstract domains of experience (e.g., psychological states, legal relationships, formal Systems) (Johnson 1991). In the cases of bodily and visual balance, there seems to be one basic scheme consisting of a point or axis around which forces and weights must be distributed so that they counteract or balance off one another. Our experience of bodily balance and the perception of balance is connected to our understanding of balanced personalities, balanced views, balanced Systems, balanced equilibrium, the balance of power, the balance of justice, and so on. In each of these examples, the mental or the abstract concept of balance is understood and experienced in terms of our physical understanding of balance. Image Schemas have internal logic or structure that determine the roles these Schemas can play in structuring various concepts and in patterns of reasoning. It is not the case that a large number of unrelated concepts (for the systematic, psychological, moral, legal, and mathematical domains) all just happen to make use of the same word balance and related terms (Johnson 1991). Rather, we use the same word for all these domains because they are structurally r

When "go" means "come": Questioning the basicness of basic motion verbs

Abstract: The purpose of this paper is to question some of the basic assumpiions concerning motion verbs. In particular, it examines the assumption that \"come\" and \"go\" are lexical universals which manifest a universal deictic Opposition. Against the background offive working hypotheses about the nature of'come\" and ''go\", this study presents a comparative investigation of t wo unrelated languages—Mparntwe Arrernte (Pama-Nyungan, Australian) and Longgu (Oceanic, Austronesian). Although the pragmatic and deictic \"suppositional\" complexity of\"come\" and \"go\" expressions has long been recognized, we argue that in any given language the analysis of these expressions is much more semantically and systemically complex than has been assumed in the literature. Languages vary at the lexical semantic level äs t o what is entailed by these expressions, äs well äs differing äs t o what constitutes the prototype and categorial structure for such expressions. The data also strongly suggest that, ifthere is a lexical universal \"go\", then this cannof be an inherently deictic expression. However, due to systemic Opposition with \"come\", non-deictic \"go\" expressions often take on a deictic Interpretation through pragmatic attribution. Thus, this crosslinguistic investigation of \"come\" and \"go\" highlights the need to consider semantics and pragmatics äs modularly separate.

A model of the human capacity for categorizing spatial relations

Abstract: Languages vary dramatically in their structuring ofspace. Despite this wide Variation, however, the search for universals in spatial semantics is well motivated by thefact that all linguistic spatial Systems are based on human experience ofspace, which is in turn constrained by the nature ofthe human perceptual System. I present a connectionist model which contributes to the search for universals in this domain. Its design incorporates a number of structural devices motivated by neurobiological andpsychophysical evidence concerning the human visual System; these provide a universal perceptual core which constrains the process ofsemantic acquisition. Using these structures} the model learns the perceptually grounded semantics for closed-class spatial terms from a ränge oflanguages—providing at least a preliminary model of the human capacity for categorizing spatial events and relations. The model gives rise to two predictions concerning the manner in which one can expect tofindmotion encoded in closed-class spatial terms in the world's languages. l. The linguistic categorization of space The linguistic categorization of space is a topic that has captured the attention of linguists and other cognitive scientists for a number of years. There is good reason for this. Spatial location is often expressed by closed-class forms, which have \"the fundamental role of acting äs an organizing structure for further conceptual material\" (Talmy 1983: 4). Thus, the use of these forms to denote spatial location is an indication that space has a privileged position äs a foundational ontological category in language, a position which most other domains do not share. This point is strengthened by the fact that space often serves äs a source domain for metaphoric understandings of other parts of the conceptual System; its influence is therefore not localized to an isolated sphere of Cognitive Linguistics 6-1 (1995), 63-88 0936-5907/95/0006-0063 © Walter de Gruyter

Language as context, language as means: Spatial cognition and habitual language use

Abstract: This paper explores the conceptual representations of spatial relationships in manipulable (or \"table-top\") space. Three cognitive tests demonstrate that subjects drawn from two Tamil subpopulations tended to mentally encode relationships among objects in a way which correlates with their different habitual uses oflinguistic spatial reference.

American friendship and the scope of metaphor

Abstract: This paper raises the issue ofthe generality of the application ofparticular source domains to particular target domains and suggests a new, theoretically useful notion for this purpose: t hat ofthe \"scope of metaphor\"; t hat is, the ränge ofthe application ofparticular source domains to target domains. The example on which the study is based is the American conception of friendship. 1t is found that there are no metaphor s (i.e., source domains) that are used exclusively for friendship. The abstract concept of friendship äs target appears to derive its source domains from six large metaphor Systems: \" communication\" metaphors, \"emotion\" metaphors, \"state\" metaphors, \"complex Systems\" metaphors, \"event structure\" metaphors, and \"positive/negative evaluation\" metaphors. If these findings are correct, metaphor research will have to turn to several new tasks in the future: (1) to determine whether the metaphor Systems suggested in this work are really there in the metaphorical System ofEnglish; (2) ifthey are, to describe fully the metaphor Systems presented only partially in this paper; (3) to isolate similar large Systems of metaphor s; (4) to describe further abstract concepts with the help of these metaphor Systems; and (5) to construct theories about how the large metaphor Systems fit together. It isproposed that in all ofthis work, the notion of \"scope of metaphor\" (or more precisely, the scope of the source domains of metaphors) plays a crucialpart.

The linguistic construction of space in Ewe

Abstract: 77/M paper presents the linguistic means of describing spatial relations in Ewe with particular emphasis on the grammar and meaning of adpositions. Ewe ( N iger-Congo ) has two sets of adpositions: prepositions, which have evolvedfrom verbs, and postpositions which have evolvedfrom nouns. The postpositions create places and are treated äs intrinsic parts or regions of the reference object in a spatial description. The prepositions provide the general orientation of a Figure (located object). It is demonstrated (hat spaiial relations, such äs those encapsulated in \"the basic topological prepositions at, in and on\" in English (Herskovits 1986: 9), are not encoded in single linguistic elements in Ewe, but are distributed over members of dijferent form classes in a syntagmatic string, The paper explores the r öle of compositionality andits interaction with pragmatics to yield understandings of spatial configurations in such a language where spatial meanings cannot he simply read off one form. The study also examines the diversity among languages in terms of the nature and obligatoriness of the coding of relational and ground Information in spatial constructions. It is argued that the ränge and type of distinctions discussed in the paper must be accountedfor in semantic typology and in the cross-linguistic investigation of spatial language and conceptualisation.

The anthropocentricity of the English word(s) back

Abstract: The semantics oflexical items manifest in the form back are closely examined and related t o front, face, behind Essentially, \"back\" is where the action and interaction isn't; i.e., it names the intrinsic pari, location, or direction directly opposite where the action and interaction goes on. By defining the back äs \"that pari of a body opposite the interactive-side\"— originally on an anthropomorphic model of the prototypical human being in upright stance confronting the world by looking forward and walking forward—the term is extended to the corresponding proper parts ofvertebrate and invertebrate animals and to inanimate objects such äs houses, cupboards, and Computers. Although some semantic extensions of \"back\" in languages other than English are based on an animal model (are zoomorphic)t the anthropomorphic model is shown to be prior. The anthropocentric \"journey Schema\" is the basis for extending back to abstracts such äs spatial-temporal location, movement, notions offalling behind one's peers, and supporting someone or something. The journey Schema underlies the \"landmark\" model for semantic extensions of'back\" in a number of languages. The paper confirms that the uses and meanings of English back are motivated by our cognitive modelling of the world and that they evidence a powerful anthropocentric image of \"the body in the mind\" ofhumankind. The principal purpose of this article is to examine very closely the semantics of lexical items manifest in the form back. It offers an explanation for the extension of the original human body-part term to other vertebrate and invertebrate creatures, to inanimate objects, to spatialtemporal location, to acts and to movement. Semantic extensions of body-part terms have been discussed for languages from many families across the world by Basso (1967: 351-352), Hill (1975, 1982), Fillmore (1982), Brugman (1983), Svorou (1988), Heine (1989), MacLaury Cognitive Linguistics 6-1 (1995), 11-31 0936-5907/95/0006-0011 © Walter de Gruyter

A coding system for spatial relational reference

Abstract: A coding System is presented for the semantic description of natural language expressions referring to spatial relations. The coding system has been developed with the principal (but not exclusive) aim of conducting crossHnguistic analyses of spatial language acquisition and development. Theoretical and methodological problems attendant upon the analysis of spatial relational meaning are addressed. The coding system is described and its use is illustrated. A reformulation is proposed of the traditional distinction between sense and reference in terms ofa notion of \"meaning äs mapping\". The specific research questions which the coding system is designed to address are outlined.

Family resemblance in the Dutch spatial prepositions door and langs

Abstract: Spatial prepositions are on the whole not amenable to a semantic description in terms of a criterial set of necessary and sufficient features. Indeed, the various uses/readings of spatial prepositions are linked through similarity rather than identity. Their conceptual/semantic structure, then, can most appropriately be described in terms of a family-resemblance network of different, yet mutually related readings. The body of this paper examines the semantic/conceptual Information in the Dutch spatial prepositions door and längs and sets up their familyresemblance structure. These, then, are the main lines of the analysis: each ofthe uses of door and längs is specified in terms ofa bündle of cooccurring features (or in other words, in terms of a featural configuration). Taken together, these uses constitute a network of interrelated featural configurations in which configurations either have features in common with one another or are transformationally linked. As the family-resemblance structures I would like to present systematically characterize the semantic/conceptual relations between the various uses o/langs and door, they also shed some light on how one use might be derivedfrom (or motivatedby) the other. L Family resemblance in cognitive semantics Cognitive semantics makes specific Claims about lexical-semantic structure. While it does not exclude that the conceptual/semantic Information in some lexical items may be analyzed in terms of a criterial set of necessary and sufficient features, it maintains that, on the whole, they are not amenable to this type of analysis. Indeed, members of a category in the extension of a lexical item may be linked not because they all share the same criterial set of attributes, but because they share different sets of attributes with each other; in other words, because they are similar to each other in different respects Cognitive Linguistics 6-2/3 (1995), 183-207 0936-5907/95/0006-0183 © Walter de Gruyter

Grammatical category and world view: Western colonization of the Dyirbal language

Abstract: Lakoff (1987) presents the noun class system of Dyirbal, an indigenous Australian language, as a prototypical example of categories conforming to his theory of categories. He takes as the title of his book a summary of the concepts encoded by nouns belonging to the so-called feminine gender of Dyirbal: women, fire and dangerous things, and suggests that the prototypical members of two of the four noun classes are human males and human females respectively. (The other two classes are described as “vegetable food'” and “everything else”.) It is argued in this paper that Lak off proposed a human sex-based analysis of these two classes, not because this analysis best fits the data, but as a result of applying western cognitive models to the analysis. If the noun category system of Dyirbal is analyzed in terms of potency and harmony, it can be seen to be more relevant to the experience and cognitive needs of the Dyirbal-speaking people than is indicated by Lakoff's analysis. The noun class system of Dyirbal is shown not to be constructed around real-world entities as prototypes, but around sets of properties which define the categories.