"A Metaphor Is Pinning Air to the Wall": A Literature Review of the Child's Use of Metaphor.

Two questions guide an examination of the literature on children's use of metaphor in their learning: How is metaphorical language used in the process of learning? And, what are the implications for teachers in helping their students to use metaphorical language and thereby build knowledge of their world?

Literature About a Child's Use of Metaphor

Much of the research in children's use of metaphor over the last 40 years has been within the context of stages of cognitive development (Asche & Nerlove, 1960; Billow 1975; Winner, Rosentiel, & Gardner, 1976). Research has been inspired, generally, by interest in the nature of creative thought (Bruner, 1986; Gardner, Kirchner, Winner, & Perkins, 1975), and, specifically, by how comprehension of metaphor is related to ideas of cognitive development (Pollio & Pollio, 1974; Vosniadou, Ortony, Reynolds, & Wilson, 1984; Waggoner & Palermo, 1990).

The metaphors that younger children use are based on physical links, rather than conceptual and psychological ones (Winner, 1982). As children become more expert in reading and writing, and as they grow older, their use of spontaneous metaphor declines (Wagner, Winner, Cicchetti, & Gardner, 1981; Winner et al., 1976). Production and comprehension of metaphor may be related to the nature of conversation between children and the teacher (or other adult). That is, children's exploration and understanding of metaphor are diminished by teachers and other adults (Gallas, 1994; Winner, 1982) who may look for specific meanings of metaphorical language. Young elementary school students are able to grasp functional and concrete metaphoric polarities well before they can explain their specific meanings (Asche & Nerlove, 1960; Cicone, Gardner, & Winner, 1980; Winner et al., 1976). Children as young as 5 and 6 can use concrete, functional metaphors, and even explain their choices (Waggoner & Palermo, 1990). Comprehension of metaphor is related to the context of the metaphorical statement and prior knowledge (Ortony, 1979), and to the context of the research situation itself (Vosniadou et al., 1984).

In their seminal study, Asche & Nerlove (1960) explored children's comprehension of descriptive "double function" terms such as "hard," "deep," and "bright." A researcher talked with each child in an elementary school and showed each child a number of objects. From among an ice cube, a wooden block, a branch, a powder puff, and a cube of sugar, the child was asked to pick an object that exemplified the word "sweet." Then, the child was asked to name objects that the term described to find out whether the child grasped the physical properties. In the next step, the child responded to the question, "Are people sweet? Do you know any sweet people?" Asche and Nerlove (1960) found that "sweet" was the only double function attribute that the children could talk about. The older children from the school "showed evidence of an increase in the use and understanding of the psychological sense of the terms" (p. 52). These children understood the concept of dual function in comments such as, "Hard things and hard people are both unmanageable."

Billow (1975) conducted studies with 50 boys, ages 5-13, recruited randomly from the files of a Long Island, New York, yeshiva. In the first phase, a researcher presented the children with groups of metaphors: "Hair is spaghetti," "The pond is his mirror," "A butterfly is a flying rainbow" (Billow, 1975, p. 416). Each boy was given two presentations of the sentences, one verbal and one pictorial. In their answers, the youngest children (age 5-7) transformed the relationship of metaphoric comparison into other types of relationships. One 5-year-old, for example, understood the sentence "The branch of the tree was her pony" (p. 416) to mean that the pony carries heavy things like the branch does. The same child explained the statement "A flower is a grounded bird" (p. 416) in terms of a contiguous relationship: "A bird is on the flower and the flower is on the ground" (p. 420). To a child of 7, the metaphor "The stars are a thousand eyes" (p. 418) became a statement of cause and effect: "It means, they light up all over, they make you see in the dark" (p. 420). The same child found a commonality between rainbow and butterfly in their both being colorful. The researcher pursued the subject, using principles of inclusion (Piaget, 1969). He repeated the child's answer and added, "Tell me, are there more colorful things in the world or more butterflies?" (Billow, 1975, p. 416).

The child responded, "Colorful things."

"Why?"

"Because butterflies are colorful and people like colorful things and have a lot of them in houses" (Billow, 1975, p. 416). The experimenter returned to the original question about why there are more colorful things in the world than there are butterflies, and the child responded, "[There are] only one or two rainbows in the world. They come back, taking turns. Rainbows never wear out" (p. 417).

In their study, Winner et al. (1976) looked at processes of comprehension. One hundred and eighty children between the ages of 6 and 14, evenly divided between girls and boys, were selected at random from their classrooms. The study had two parts. In the first, the researcher read a story to the children after explaining, "I want you to tell me what you think the story means." In the second part, the children were read a story and were asked to choose among four different things "that some people think the story might mean" (p. 292). Of the 16 metaphoric sentences embedded in each test, eight were cross-sensory (e.g., "her perfume was bright sunshine") (Gardner et al., 1975). Eight were physical-psychological (e.g., "the prison guard was a hard rock") (Asche & Nerlove, 1960). Winner et al. (1976) found that the youngest children responded with magical interpretations; that is, they offer "a paraphrase ... that maintains the literal meaning" (Winner et al., p. 293), such as "the king turned the prison guard into a rock" (p. 293). The older children were more likely to make genuine metaphorical choices, such as, "The guard was mean and did not care about the feelings of the prisoners" (Winner et al., 1976, p. 295).

Vosniadou and Ortony (1983) found that young children discriminated among literal, metaphorical, and anomalous similarities of 10 nouns embedded in statements read to them. Vosniadou et al. (1984) looked more closely at the children's processes of thinking and talking about the metaphors presented to them. Ninety children (half of them girls and half of them boys) from a nursery and an elementary school in rural Illinois took part in the study. The children enacted the endings to stories presented to them in which the choices of outcomes were stated in metaphoric terms. The children acted out the meanings they made of the statements. "Acting out the entire story forces children to process the story's content, making it more likely they will use the content to understand the metaphor" (Vosniadou et al., 1984, p. 1591). The authors found that the children's comprehension was not dependent on the presence of a metaphorical statement, but rather was determined primarily by the nature of the story context in which the endings were given. Second, their comprehension depended on the extent to which words, especially verbs, were literal ("Billy was a child hiding the cookies") or nonliteral ("Billy was a squirrel burying the nuts") (Vosniadou et al., 1984, p. 1591). One child "made Billy hide the cookies, but tried to enact burying literally" (p. 1593).

Waggoner and Palermo (1990) challenged the conclusion that young children do not comprehend psychological metaphors, stating that some descriptions of personality traits, such as bossiness and generosity, "may be too abstract for young children to understand at the ages tested" (p. 152). Instead, Waggoner and Palermo focused on children's perceptions of emotional states. The authors believed that children as young as 3 or 4 would be familiar with feelings such as sadness, anger, fear, happiness, love, and hate. They presented 48 metaphors in 12 stories, contrasted in pairs. Children performed better on what the authors considered negative emotions, such as anger, sadness, fear, and hate, than they performed on metaphors for positive emotions, such as happiness and love. The children differentiated among incorrect metaphors and did not simply choose an incorrect figurative answer for their psychological answer. For instance, although the researchers considered both "sorrow" and "fear" to be incorrect ways to explain the metaphor "Joe is a snorting bull" (p. 158), the children did not treat the choices equally. They were more likely to interpret the metaphor as representing fear. The authors conjectured that the children were thinking of the fear they would experience when facing a snorting bull, rather than considering anger as an attribute of the bull. The authors noted:

Anger metaphors were more difficult [for the children to comprehend] than fear metaphors. Although they had difficulty with the anger metaphors when they were in stories about anger and fear, the children had no such difficulty with anger metaphors when they were in stories about anger and sorrow. Sorrow metaphors were correctly interpreted in the context of sorrow versus anger but poorly interpreted in the context of sorrow versus fear. (p. 158)

Waggoner and Palermo (1990) found that the anger-fear pair was the most difficult one for children of all ages to understand, while love-hate metaphors tended to be the easiest. The anger metaphors contained words or concepts that young children found frightening, such as "buzzing bees," "stinging words," and "a bucking horse." The children interpreted a passage about a boy feeling like a sinking boat to mean something fearful, much as they responded to the snorting bull metaphor. "This tendency might be associated with the child's response if placed in a situation of facing a snorting bull or sitting in a sinking boat" (p. 161).

In the Waggoner and Palermo (1990) study, the children related what is familiar--their conception of feelings they would have in a difficult situation--to what is unfamiliar (a sinking boat and a snorting bull). Their interpretations are direct and concrete, and centered in their immediate experiences. In the more difficult anger-fear pairs, concrete metaphors were easier to explain than abstract ones. Concrete metaphors make the strange familiar. To say that one dinosaur was the height of 14 men standing on one another's shoulders would be comprehensible to most children. Children need such comparative attributes to comprehend a metaphor.

Karen Gallas (1994) found that her 1st-grade students, given free rein to explore such questions as "What is the beginning of nature?" could create their own metaphors and make their own connections between the familiar and the strange. Watching and listening to children's conversations, rather than prompting them, could reveal a new understanding of children's use of metaphor (Gallas, 1994; Winner, 1982). A teacher in this situation must be patient, for without a doubt, the children's exploration process is a scattered one. "To the adult listener, the trains of [children's] thought are often difficult to follow and give the impression the talk is going nowhere" (Gallas, 1994, p. 105).

How Is Metaphorical Language Used in Learning?

For children to gain fluency in any given subject, they need practice in speaking, not just in listening. "It is when we have to put words together and make sense, when we have to formulate questions, argue, reason, and generalize, that we learn the thematics of science" (Gallas, 1994, p. 24).

The ways we expect children to talk, think, and write about science make a large assumption about what the language of science is and ought to be. We ask, "What is a shadow?" seeking a generalizable answer that will show that children have grasped the overall theoretical point. The children reply, "It's part of darkness." "It's night lying down." Another responds, "Day is night time for the shadow." They speak and write metaphorically, in terms of the particular. (Gallas, 1994, p. 97)

Wolf (1982) suggests that the creation and expression of metaphorical language does not follow a simple, linear progression through development, but includes recent events, pervading traits, and motivations that characterize the child's life. "The distinctiveness of a child's product reflects the interaction of several factors" and "do not stem from any single dimension, nor from a particular level of symbolization" (p. 121).

Gallas (1994) writes:

the object of teaching science is not to teach the correct usage and application of scientific concepts and terminology, but rather to engage as many children as possible in observing, experimenting, talking, and writing about the world. That process must begin with their deep emotional attachments and focus, as children naturally do, on the surprises in nature. (p. 98)

The teacher focuses less on the answers the children come up with than on the struggles they undergo together to create arguments that offer evidence from their experience. Children's "language does not parrot the forms that we believe indicate real mastery of subject matter" (Gallas, 1994, p. 97).

Implications for Teachers in Understanding Children's Perceptions and Knowledge of Their World

For an utterance to count as metaphoric, the speaker must know the conventional extension of the word used, such as "tail" for "string," or "scarecrow" for "pencil." The speaker knows that items in two different domains are related. If this condition is met, the child who calls the string a tail or the pencil a scarecrow has established an unconventional link between word and object; hence, "it can safely be inferred [that] this non-literal designation is intended" (Winner, 1979, p. 475). When a child notices a catch comb on the front of a train in a photograph, for instance, and calls it a broom sweep, her reference is not a metaphor unless she knows the function of the catch comb. At this point, the adult should pursue the child's line of thinking: What is the child's perception of the catch comb? Why is it a broom sweep? That is, what does the child think about the connections she makes?

Pursuing the line of thought is as important as the answer, or metaphor, derived from the child's thinking. Both Winner (1982) and Gallas (1994) found that teachers and other adults often undermine children's verbal exploration of metaphor. Winner (1982) recounts a conversation during a Seder dinner, in which the man conducting the ritual meal described how, after one of the plagues, Pharaoh's heart had hardened to stone. The children looked bewildered, and he asked them what they thought the phrase meant. The youngest, a child of 5, supposed God had come down and literally turned Pharaoh's heart to stone. A 6-year-old disagreed, speculating that Pharaoh lived in a castle made of stone. An 8-year-old interpreted hardness as having to do with Pharaoh's muscles. "Eventually, an adult suggested that Pharaoh's mood, rather than his heart or his muscles, had changed" (Winner, 1982, p. 159).

Winner (1988), Palermo (1986), and Vosniadou et al. (1984) questioned the extent to which prior knowledge of psychological states described metaphorically affects metaphoric comprehension. A child's failure to interpret the metaphor "the prison guard was a hard rock" (Winner et al., 1976) might indicate unfamiliarity with prison guards, or the quality of being called hard. Children may not understand the psychological abstraction of character, but they may understand metaphorical attributes of character in other contexts (Vosniadou et al., 1984). Pearson, Raphael, Tepaske, and Hyser (1981) discovered that children understood unfamiliar information better with the use of metaphor than they did familiar information, yet metaphor seemed to impede comprehension of familiar information. This finding suggests children use metaphorical language to understand the world and to build knowledge as much as they develop it as a result of their knowledge.

Teachers may not be expert in creating metaphor, but they need to speak metaphorically in order for students to learn to use metaphors. Teachers can model their thinking processes in figuring out ways to interpret what they hear, read, and see. Awareness and use of metaphor is cumulative. Practice in creating and using metaphor in conversation and writing leads to new questions and perspectives.

Conversation and writing about experience and literature help to create context for a metaphorical statement. Vosniadou et al. (1984) note, "In real life, children are not usually exposed to metaphors out of context" p. 1589). Empirical studies, out of context, might create problems as much for adults as they do for children. The belief that texts containing metaphors are more difficult to read than texts without them is not necessarily simply a matter of the presence of the metaphor. Where metaphors are forced, contrived, or out of context, text is less comprehensible, not because of the metaphors but because of a lack of coherence.

Some might say that a teacher changes the definition of metaphor to suit the context, rather than to explore how dialogue fits a definition of metaphor. The developmental, cognitive view of the use of metaphor found in the literature begs the question of how children reveal their understanding and their perceptions. How do children formulate their understanding of new information? When the children in Karen Gallas's classroom talk about the question, "Does the universe end?," one of them asks, "Do you mean, um, like, will it like be gone some day, or do you mean, does it go on out in space all the way?" (Gallas, 1994, p. 109). That the production of metaphor may be as natural as speaking suggests that a teacher can promote, through discussion, an environment where children are free to explore, no matter how disordered the process may be.

References

Asche, S. E., & Nerlove, H. (1960). The development of double function terms in children: An exploratory investigation. In B. Kaplan & S. Wapner (Eds.), Perspectives in psychological theory: Essays in honor of Heinz Werner (pp. 47-61). New York: International Universities Press.

Billow, R. A. (1981). Observing spontaneous metaphor in children. Journal of Experimental Child Psychology, 31, 415-445.

Bruner, J. (1986). Actual minds, possible worlds. Cambridge, MA: Harvard University Press.

Cicone, M., Gardner, H., & Winner, E. (1980). Understanding the psychology in psychological metaphors. Journal of Child Language, 8, 213-216.

Gallas, K. (1994). The languages of learning: How children talk, write, dance, draw, and sing their understanding of the world. New York: Teachers College Press.

Gardner, H., Kirchner, M., Winner, E., & Perkins, D. (1975). Children's metaphoric productions and preferences. Journal of Child Language, 2, 125-141.

Ortony, A. (1979). Beyond similarity. Psychological Review, 86(3), 161-181.

Palermo, D. (1986). Metaphor: A portal for viewing the child's mind. In L. P. Lipsitt & J. H. Cantor (Eds.), Experimental child psychologist: Essays and experiments in honor of Charles C. Spiker (pp. 111-137). Hillsdale, NJ: Lawrence Erlbaum.

Pearson, P. D., Raphael, T. E., Tepaske, N., & Hyser, C. (1981). The function of metaphor in children's recall of expository passages. Journal of Reading Behavior, 13(1), 249-261.

Piaget, J. (1969). Language and thought of the child. Cleveland, OH: Meridian Books.

Pollio, M., & Pollio, H. (1974). The development of figurative language in children. Journal of Psycholinguistic Research, 3(3), 185-201.

Vosniadou, S., & Ortony, A. (1983). The emergence of the literal-metaphorical-anomalous distinction in young children. Child Development, 54, 154-161.

Vosniadou, S., Ortony, A., Reynolds, R. E., & Wilson, P. (1984). Sources of difficulty in the young child's understanding of metaphorical language. Child Development, 55, 1588-1606.

Waggoner, J. E., & Palermo, D. S. (1990). Betty is a bouncing bubble: Children's comprehension of emotion-descriptive metaphors. Developmental Psychology, 25(1), 152-163.

Wagner, S., Winner, E., Cicchetti, R., & Gardner, H. (1981). Metaphorical mapping in human infants. Child Development, 52, 728-731.

Winner, E. (1979). New names for old things: The emergence of metaphoric language. Child Language, 6, 469-491.

Winner, E. (1982). The child is father to the metaphor. In H. Gardner (Ed.), Art, mind, and brain: A cognitive approach to creativity (pp. 158-169), New York: Basic Books, HarperCollins.

Winner, E. (1988). The point of words. Cambridge, MA: Harvard University Press.

Winner, E., Rosentiel, A. K., & Gardner, H. (1976). The development of metaphoric understanding. Developmental Psychology, 12(4), 289-297.

Wolf, D. (1982). Max and Molly: Individual differences in early artistic symbolization. In H. Gardner (Ed.), Art, mind, and brain: A cognitive approach to creativity (pp. 110-129). New York: Basic Books, HarperCollins.

Author's Note: During a conversation about metaphor in Teacher As Writer, A Writing Seminar, Evelyn Beaulieu, Julie Orton, and Brenda Power thought of the title metaphor.

Sandip Lee Anne Wilson is a doctoral student, College of Education and Human Development, University of Maine, Orono.