ABSTRACT
Drawing on Bakhtin (1984), this study explored multivoicedness in preschoolers’ and young schoolchildren’s play. Multivoicedness could be seen in preschoolers’ and schoolchildren’s differentiation between pretend and nonpretend voices: for example, in their regulation of tempo, loudness, and voice. These findings are in line with prior research in that several studies have demonstrated children’s fine differentiation between pretend and nonpretend voices and between different pretend role voices. Similarly, this study corroborated earlier findings on a socialization dimension (Auwärter, 1986; Sawyer, 1996) in that schoolchildren were more likely than preschoolers to stay within the pretend frame, without talking out-of-frame in a nonpretend voice. Last, and most important, the play dialogues showed how children reproduced collusion in adult–child talk in enacting doctor–parent–child talk. Such collusion was seen in the children’s subtle ways of employing collusive forms of production formats, such as the collaborative “we”-form when playing doctor (talking to a sick child). On the basis of these findings, it is argued that voice cannot merely be equated with “role” or roles. In employing a collusive production format, the children demonstrated a Bakhtinian awareness of subversion in adult– child talk. On a theoretical note, the present findings support voice models rather than frame models of children’s play. Also, extending the work of Göncü (1993), it is proposed that children’s collusive voices can be seen as an important type of tertiary intersubjectivity.
