Writers are also readers, and they write in the shape of their readerly desires. David A. Brewer's study of “imaginative expansion” (2) explores one interface between such reading and writing: eighteenth-century British “fan fictions” that imagine further adventures (such as backstories and sequels) for beloved characters. Brewer offers a history of reading from below and a novel take on popular literary history. Taking “popular” in the sense of “popularity,” he focuses on the “social canon” (17), those books selected by readers when they are not reading for prestige or improvement, but for pleasure and even company. By taking seriously such naive and sentimental reading—the playful belief that characters exist off the page and can be more deeply known—Brewer's “internal” history of reading supplements “external” histories that focus on inventories of reading materials but cannot fully address the way that they are used (7–8).

Brewer's account begins with the emergence of “imaginative expansion” in the early eighteenth century, which was an effect of relative political stability. As party strife lessened, habits of political reading (such as reading à clef) gave way to more sociable kinds of reading. Popular texts like Gulliver's Travels (1726) and The Beggar's Opera (1728) come to seem more fictional than satiric, and as they survive their local references and lose their topicality, they spawn additional works about their characters in a “Fictional Archive” that supplements and enriches the originals (chap. 1). Brewer focuses on two ways that readers extend and deepen texts: “Visualization” (chap. 2) and “Character Migration” (chap. 3). Richard Steele's sketch of Inkle and Yarico in the Spectator (1711) was expanded by rewritings that more fully visualize the tragic parting and testify to readers’ desires for more information about these characters, or simply for more of them. Similarly, Falstaff migrates to other texts that add adventures, complete his backstory, and claim him from the “textual commons” (84) for their own use. Such further uses are integral to the feedback loop of “social canonicity” (95): the greater the number of textual sites, the greater the sense of a character's inexhaustibility. Falstaff, Roger de Coverly, and Quixote, for instance, transcend their textual sources, seeming to be “everywhere and nowhere” (80) and so available for further exploration. And as such “old friends” (105) take on lives of their own, they serve as the locus of virtual communities, a neat way to trace links between sociable reading practices and the new kinds of sociality available in print culture.

The second half of Brewer's study examines the increasingly successful attempts of authors to exploit this mode of reading. In various ways (which Brewer cleverly explicates) Samuel Richardson's Pamela (1740) and Laurence Sterne's Tristram Shandy (1760–67) are written to harness this kind of reading. Pamela depends on the authenticating device of the heroine's epistolary archive, the depth of which is suggested by the “editor” but also policed by the author. Richardson invites readers to extend and visualize his character, but only insofar as they agree with his conception of her; the reading public becomes a coterie. Tristram's book is likewise full of invitations to help construct the story as well as examples of characters imagining further (and sentimental) adventures of others. But while readers are invited to participate, the novel is managed by Sterne on the model of a club; you give up some autonomy in order to be entertained by the master of ceremonies. By writing in a readerly mode and conflating his own identity with that of his characters, the author himself achieves the popularity and “placeless omniscience” (183) of characters like Gulliver and Falstaff. The next and final step in this development is Sir Walter Scott, who draws down the curtain on this practice of reading by jealously claiming proprietary control over his characters and their fates. This state of textual ownership, Brewer asserts, has informed conceptions of authorship and reading ever since.

Brewer narrates his story as a progression from readers freely imagining expansions through increasingly sophisticated and successful attempts by authors to control their characters, a dwindling of the spaces of reading from commons through coterie, club, and market. But this argument is not precisely a historical one. Brewer conducts his discussion of the “textual commons” through texts that postdate Pamela and are contemporary with Sterne. These various works, that is, coexist and suggest less a history of changing practices than an account of the range of kinds of reading available at a given moment. Readers could visualize Yorico's tragic betrayal, enjoy Falstaff's migration into new stories, and participate in Richardson's or Sterne's circles, all in the same week. My point is not that Brewer is wrong, but rather that his topic and materials are so compelling they finally threaten to break the frame that organizes them. Histories of reading inherently challenge notions of periodization (reading is a practice of anachronism), and I wonder if it is necessary to present this mode of reading as a rise and fall. It equally explains the recycling and adjustment of favorite materials in texts from The Golden Ass (ca. 170) through Orlando Innamorato (1495), Orlando Furioso (1516), and Don Quixote (1605, 1615), as well as Rebecca and Rowena (1849) and Wide Sargasso Sea (1966). (William Makepeace Thackeray's sequel to Ivanhoe, indeed, suggests that the period Brewer puts to his account could be a comma.) In explicating this practice, Brewer makes a valuable contribution to our histories of reading. In doing so, he also may suggest more than his historicism can bear, a readerly dynamic defined less by any particular historical location than by the way it enables the anachronisms that allow readers at once to participate in their histories and to expand them imaginatively.