Jeffrey Eugenides's 2002 Middlesex, a critically acclaimed historical novel, has been praised as an expansive, epic portrait of the American twentieth century from its immigrant roots to the present. (1) It takes its readers from a Turkish village in the 1920s to the race riots of the late 1960s, following a Greek and then Greek American family across time and the world, spinning an interestingly twisted yarn in the voice of the family's latest product, whose gender identity, complicated by a genetically inherited hermaphroditism, is at the center of his story. The novel displays a particular historical imagination, as all historical novels do; it depends on a set of notions about the relationship between past, present, and future, about cause and effect, and about the possibilities and problems that attempts to understand and represent the past entail. And, as is also the case with all historical novels, its historical imagination can tell us something about the historical imagination of its times. While it is no longer common in current critical discourse to discuss works in terms of aesthetic failure or success, I believe that Middlesex fails aesthetically and that it is important to talk about it in these terms because how it fails says something about its historical imagination and the historical imagination of its times. At the root of its failure, I'll argue, is the way it imposes a false closure on its narrative of the main character's gender crisis. This closure represents something other than a poor aesthetic choice. Rather, its falseness--the unearned, unwarranted character of the novel's ending--is unintended, and so it represents a failure that is especially indicative of the unconscious effect of its historical imagination. The way Middlesex ends is in part due to, and thus can tell us something about, the way history felt in America in 2002. The formal phenomenon of closure is closely linked in the literature and thought of the second half of the twentieth century to the existential phenomenon of contingency. The felt relation of a time, a writer, or a particular kind of historical imagination to the fact of contingency informs the way stories are told, in particular the nature of their endings (and not just in the recent past, as Frank Kermode has shown). Postmodernism, though, has been especially invested in the connection between closure and contingency. In an early (1972) statement of postmodern doctrine made in the first issue of boundary 2, the postmodern journal he cofounded, William Spanos describes the relationship between closure and contingency after midcentury: Only after the existentialist philosophers revealed that the perception of the universe as a well-made fiction, obsessive to the Western consciousness, is in reality a self-deceptive effort to evade the anxiety of contingent existence by objectifying and taking hold of it, did it become clear to the modern writer that the ending-as-solution is the literary agency of this evasive objectification. (152) The distrust of closure is widely articulated in postmodern thought, early (as in statements by Leslie Fiedler, Ihab Hassan, John Barth, et al.) and late (Linda Hutcheon, Fredric Jameson, Brian McHale, et al.). The story that I will tell through my reading of Middlesex, however, is that of the changing nature of that distrust in recent years. What endings mean, and why writers embrace them or avoid them, depends in part on how contingent existence feels and how public discourse and constructions of history deal with that feeling. As a result, events that reawaken a sense of contingency and challenge already constructed narratives--in particular, historical traumas--can affect the shape of literary endings. This is especially the case, I will argue, for historical literature, work whose focus is explicitly on the past and always implicitly, as a result, on the way that history ends--on the way the past leads to the precarious present and, ultimately, the future. …