Metaphor: Theoretical and empirical research.

Abstract

Metaphor plays a major role in our understanding of language and of the world we use language to talk about.Consequently, theories of language comprehension and of language itself are incomplete if they do not handle the phenomenon of metaphor, and they are inadequate if they cannot.Traditional definitions and theories of metaphor are reviewed. It is suggested that they err in equating metaphors with comparisons rather than merely implicating comparisons. Empirical research is reviewed, revealing, for the most part, serious problems, particularly in the developmental research. These problems often relate to inadequate underlying theories about the nature of metaphor. Other difficulties include inadequate controls over preexisting knowledge, and overly hasty conclusions that children cannot understand metaphors. Related research on the comprehension of proverbs and analogies is discussed. Some recommendations for future research are made. These depend on a redefinition of metaphor and on the employment of an investigative approach that will permit adequate controls of preexisting knowledge, surface structure, and meaning.The approach recommended emphasizes and takes advantage of the contextdependent nature of metaphors. Finally, the role of comparisons is reexamined. It is of no avail to argue that metaphors are really implicit comparisons if, in so doing, one hopes to account for or explain their nonliteral nature. For even if metaphors can be transformed into comparisons, these comparisons are themselves nonliteral, and consequently still need to be explained. happens that speakers or writers do not intend what they say to be taken literally; if it is taken literally, it often either makes no sense at all with respect to the surrounding context, or it appears to express something that is either impossible or false. It is our thesis that if research into the comprehension of natural language is to have ecological validity, it is necessary that we go beyond the more traditional research goals, and extend our investigations to nonliteral uses of language, which, together with literal uses, constitute the whole range of linguistic communication.One class of nonliteral uses of language that has recently started to receive attention from psychologists is "indirect speech acts." This classis not of primary concern here (but see Clark & Lucy, 1975;Schweller, Brewer & Dahl, 1976). Rather, we are concerned with metaphor and related figurative uses of language such as similes and, to some extent, analogies. Throughout our review we use the term metaphor somewhat loosely; sometimes we use is narrowly to refer to specific utterances that are, or contain metaphors in the usual sense of the word; sometimes we use it more broadly to refer to related nonliteral uses of language such as Metaphor 3 similes or analogies.Extant theory and research concerning the development of the production and comprehension of metaphor yields numerous opinions and contradictory findings. For example, one group of studies (characterized by...

Language and the career of similarity

“…We emphasize instead the logical dependency of higher-order predicates on prior possession of their lower-order arguments. We therefore leave open the possibility that the progression may be governed by the degree of knowledge rather than by the child's stage of cognitive competence (e.g., Brown, 1989;Ortony, Reynolds, & Arter, 1978).…”

“…However, we differ from most prior theorists in an important respect. Rather than seek to explain the development of similarity in terms of global stages of competence, we will ask whether a weaker explanation will suffice, namely, accretion of knowledge (Brown, 1989;Brown & Campione, 1984;Gentner, 1977a, b;Ortony et al, 1978). We will return to comparisons with other views after elaborating our position.…”

Comparison and Categorization in the Development of Relational Similarity

“…Finally, recent research using looking-time paradigms has suggested that even young infants are sensitive to identity relations (Tyrrell, Stauffer, & Snowman, 1991). These kinds of results have led most researchers to adopt the knowledge-change view, that whether children can recognize relational similarity in a domain depends chiefiy on their knowledge of that domain (Brown, 1989;Brown & DeLoache, 1978;Brown, Kane, & Echols, 1986;Carey, 1985Carey, , 1991Chen & Daehler, 1989;Chi, 1981;Chi, Feltovich, & Glaser, 1981;Crisafi & Brown, 1986;Gentiier, 1977aGentiier, , 1977bGentiier, , 1988Gentner & Rattermann, 1991;Goswami, 1991Goswami, , 1992Goswami & Brown, 1989;Ortony, Reynolds, & Arter, 1978;Vosniadou, 1987Vosniadou, , 1989.…”

The Function of Metaphor in Children's Recall of Expository Passages

“…First, other studies have suggested that figurative language expressions may be comprehensible to children as young as first grade in certain conditions (e.g., Gentner, 1977;Home, 1966;Mayer, 1975;Pollio & Pollio, 1974;Reynolds & Ortony, Note 1). Second, Ortony (1981) and Ortony, Reynolds, and Arter (1978) have criticized much of the metaphor research on methodological grounds, arguing, for example, that the research has not adequately controlled for response bias or world knowledge. With respect to response bias their argument is that children may simply choose to respond literally even when they do in fact understand the metaphor.…”