This article responds to Keith Busby's call for a codicological approach to medieval French literary history and genre, and also seeks to demonstrate the importance of what Busby calls the geography of the codex. The Grundriss der romanischen Literaturen des Mittelalters, it is suggested, contents itself, along with most other scholarship, with a traditional account of literary history and genre grounded in a hypothetical, largely nineteenth-century, account of the chronology of textual production, one that is based largely on supposition and which pays little or no attention to manuscript dissemination. Focusing on the Roman d’Alexandre, the Roman de Troie and early romance generally, this article argues that a return to manuscripts gives a different sense of the relative importance of key texts in the evolution of the genre, and indeed of the parameters of the genre itself, at least in terms of medieval reception. This in turn engenders some reflection on the theoretical underpinnings of the Grundriss.