Abstract The repetition of a text in a medieval manuscript is rarely remarked upon as anything more than the product of scribal oversight. This essay argues that it is much more. Taking as a central exemplum the codex now cataloged in the holdings of the Bibliothèque Nationale de France as MS Français 185, the author will show how the reappearance of previously copied Lives in a single codex was received by medieval readers neither as a copyist error nor as an accident of rebinding but rather as a highly legible roadmap to the political and devotional structure of the text en mouvence. The designation in this manuscript in particular of five vitae, marked from the outset for repetition, offers precious information on the polyphonic complexities of medieval authorship, even as it provides new insight into what is perhaps the most persistent mystery in literary hagiology, that of the pleasure that might originally have undergirded a tradition so overwhelmingly constructed around the principle of repetition itself.
Abstract Compilers of manuscript verse miscellanies in early Stuart England regularly organized poems into loose clusters based on genre or topic, a practice derived from the compilation of commonplace books, where owners entered reading excerpts under topical headings. But there is another kind of clustering in miscellanies that merits attention. Several miscellanies contain verse pairings and clusters linked not by the poems’ main topics but, rather, by a single striking metaphor or image that they share, one that may be buried deep in the verse. Blowing dust in poems by Robert Herrick and Ben Jonson, for instance, leads to their pairing in six miscellanies. Jewel images also link poems internally and allude to the gift giving behind verse exchange. Metaphors of Blackness, Black beauty, Moors, and Æthiopes link other clusters and raise questions about how compilers understood race and colonization. The article charts some of the more interesting examples of this compilation convention and argues that the practice worked as a kind of coded game between compilers and readers, highlighting the compilers’ impressive familiarity with contemporary verse, readers’ keen ability to recognize the logic behind the compilation, and the way linked metaphors reveal manuscript culture’s shared values, privileges, and prejudices.
Abstract This article explores how Nabokov articulates a specific concept of memory as a curated collection of textual and material fragments in the final version of his autobiography. Rather than presenting a linear narrative, Nabokov foregrounds the curatorial nature of recollection, positing remembering and compiling as analogous processes that are predicated on retrieval, selection, and arrangement. In the final version of the autobiography, the metaphor of compilation operates beyond a thematic level, shaping the very form of the autobiographical text. Through added intertextual citation, Speak, Memory draws attention to itself as a multi-voiced compilation of intergenerational memory, which draws on and recontextualizes external sources. The integration of the added paratextual elements—photographs, map, index—extends the metaphor of compilation further into the book as a material object. The article argues that these editorial gestures complicate notions of a single authorship of memory and of the past. In this context, the metaphor of compilation assumes a critical function, repositioning the author as compiler—one who continuously reorders fragments of the past, striving for but never attaining a stable, permanent structure.
Abstract Since the release of ChatGPT in November 2022, large language models (LLMs), once the purview of machine learning experts, computational linguists, and cultural analytics scholars, have entered the public discourse. Many different metaphors have been used to describe these models, and all of them reach past thinking of large language models as simply instruments for using language (or, as the companies that sell chatbots would have it, as artificial intelligence itself)—instead, many scholars ask us to consider the model as something that both uses language and is made of language. This essay will contextualize the history of LLMs within a larger history of compilation in text technologies and the digital humanities. In this broader historical view, it is possible to understand LLMs as programs that use compilations as training inputs to produce compilations as readable output but are not themselves compilations within their model architecture. The metaphor of compilation is useful, alongside the many other metaphors used to understand these often-opaque models, as a way of capturing the historical continuities in how language models are trained and how their outputs are read.
