UNDERSTANDING VERSUS DISCRIMINATING NONLITERAL UTTERANCES - EVIDENCE FOR A DISSOCIATION

According to Olson (1988), full comprehension of nonliteral language must include the recognition of the implied meaning of the utterance and an awareness that there are two distinct levels of meaning-what is said and what is meant. Three experiments were performed to investigate children's comprehension of nonliteral language and their awareness of the say-mean distinction in such language. In Experiment 1, children heard irony modeled and were asked to produce similar utterances. Six-year-olds produced literal insults, showing that they understood the modeled irony but were not aware of the say-mean distinction. In Experiment 2, 6- and 7-year-olds were tested for awareness and comprehension of both metaphor and irony. Children were asked to distinguish nonliteral from literal utterances. There was a clear increase in awareness after the age of 6 for both metaphor and irony. However, awareness occurred independently of comprehension for metaphor, but not for irony. To determine the effect of sarcastic intonation on awareness and comprehension of irony, we designed Experiment 3 to present irony delivered in a neutral intonation. When the irony was presented without marked intonation, comprehension declined, and awareness of the say-mean distinction in irony occurred independently of comprehension, just as for metaphor in Experiment 2. Taken together, these results show that awareness of the say-mean distinction for metaphor and irony emerges after the age of 6 and that intonation facilitates comprehension of irony but does not affect awareness of the say-mean distinction. Thus, the recognition that irony and metaphor are special ways of using language does not emerge until after the age of 6.