Children's Literature Association Quarterly

In discussing the impossibility of children's fiction, Jacqueline Rose bases her case on "Peter Pan," for, as she puts it, this "is the text for children which has made that claim most boldly"—the claim being that "Peter Pan … speaks to and for children, addresses them as a group which is knowable and exists for the book, much as the book [so the claim runs] exists for them" (1). But, as she also says, "Peter Pan is that text which most clearly reveals it [the claim] as a fraud … Peter Pan has never … been a book for children at all" (1), and goes on to suggest, famously, "the impossibility" of all children's fiction. Whilst I agree with much of her thinking—namely, that the child of children's fiction is a construct; that it is presented as innocent, pure, and asexual, as a fetish allowing adults to disavow their own lack of completeness; that it is also seen as standing outside the general slipperiness of language and problems of identity; and, consequently, that it is impossible for any children's book to speak to and for children as a group—her next step, that children's fiction is thereby impossible, seems a non sequitur.

In this article I want to return to Rose's work for two reasons. First, to get behind that rallying cry of her subtitle—"the impossibility of children's fiction," which tends to be treated either as a truth to be universally acknowledged or else as refutable simply by gesturing to the humanist child (see, for example, Chapleau; Hollindale, "Introduction," "Select bibliography"¹; Lesnik-Oberstein, Children's Literature: Criticism and the Fictional Child; Children in Culture; Rustin; Walsh; Watson)—by briefly re-examining the evidence on which Rose bases her case, noting its own historical positioning. In particular, I shall suggest that she herself, ironically, holds on to a residual notion of the Romantic child, in that children's fiction is only really impossible if we see children as [End Page 290] distinct from adults, standing outside society and language, rather than being actively involved in negotiating meaning. Secondly, I will suggest that children's fiction is more viable if we adopt a Bakhtin-inflected approach, which sees the area's whole development, including "possible" readers, in dialogical terms (Bakhtin Dialogic).

Let me start, however, by examining some of the evidence Rose draws on in order to substantiate her claim. She uses "Peter Pan" for reasons given at the opening of this paper, claiming that it "has been almost unreservedly acclaimed as a children's classic for the greater part of this century" (4). However, although she repeatedly refers to this classic in italics, as though it were a single text, as "the text for children" (quoted above) and as "a children's book" (7), she also likes to keep its precise signification vague and often seems to refer to a whole body of texts, some of them not even by Barrie (for this reason, I have used the more common convention of quotation marks unless discussing specific titles²). This said, the title of Rose's monograph suggests that she is not generally alluding to the play as such (which, like most pantomimes, is ostensibly addressed to the whole family) but the "book for children" (1).³ Yet this unsubstantiated statement has, in fact, often been challenged; for instance, John Rowe Townsend's standard history of children's literature, which Rose quotes elsewhere, explicitly states that it is "not a very good book" (107), tellingly noting that "the idea of a boy who never grows up" is probably not "as appealing to children as it is to parents" and accusing Barrie of "winking over the children's heads to the adults" (106). Thus, the notion that Barrie's text "most boldly" makes the claim that it "speaks to and for children"—indeed, that it "exists for them"—has itself been subject to contestation. Earlier criticism may not have followed Rose's poststructuralist line, but it has certainly not gone unrecognized "that there might be a problem of writing, of address, and of...