a book is a fragile creature, it suffers the wear of time, it fears rodents, the elements, clumsy hands. If for a hundred and a hundred years everyone had been able freely to handle our codices, the majority of them would no longer exist.

Umberto Eco, The Name of the Rose

I returned to The Name of the Rose after hearing the sad news of Umberto Eco’s passing. For me, Eco captured in his novel our love of books and the intrigue of a mystery to be solved, of the “lost” historical text set within in a labyrinth of a medieval monastic library. His portrayal of particularly unwelcoming and hostile librarians may or may not (depending on your experience) be a work of fiction, but the novel’s suggestion of a great library now lost is unfortunately an accurate one, and it made me miserable for the many books and texts now lost to us.

The majority of medieval codices, along with the libraries that once housed them, are indeed now gone.The survival rate of medieval English texts has been estimated to be at around 2 to 5 per cent [Bale 2014, 131]. These numbers, derived largely from the examination of pre-reformation library catalogues, are problematic, for how can we accurately calculate that which is missing? But they nevertheless allow us to contemplate the potential scale of books lost to us as modern researchers, and the reasons for why this may be the case.

Certainly, many books were lost in the reformation, but this would not have had such an effect on romances, which were largely uncontroversial to over-zealous reformists. It is more likely that the Middle English romances were lost as a result of the ‘wears of time’, or ‘clumsy hands’, as Eco suggests. Middle English poems were frequently produced in small, individual booklets made of paper – less durable than the more expensive parchment – and so were far more exposed to damage, particularly if left unbound, or handled with too much enthusiasm by readers.

Or perhaps these texts were just not considered worth keeping. The Percy Folio (British Library MS Additional 27879) was supposedly first discovered being used by housemaids to kindle a fire. This manuscript is witness to several unique copies of Arthurian romance in Middle English, and without Percy’s intervention, they would have gone up in smoke and resulted in a entirely different conception of popular Arthurian romance in late-medieval England.

It is not just the missing books that we have to consider, but also the missing pages within surviving ones. The integrity of a book which is lacking half its pages presents a particular problem when attempting to reconstruct the function and meaning of a text; the text is given meaning not only by its materiality (its production materials, presentation, format, illustrations etc.), but by its textual situation; by the other texts it is presented with and read against. Likewise, in a wider sense, books can only be read against other books. If we are to get a sense of the literary culture of any given period, that which is missing has to be of equal importance to that which survives.

The low rate of survival for medieval books affects the value of those that have survived the test of time. Medieval manuscripts are valuable because of their historicity, but also because of their rarity. For example, we may deem a text particularly important because only one copy of it survives, when this copy may in fact be a representative of a widely transmitted and much read text. This text may have survived poorly because its contemporary readers either didn’t care much to keep it, or the very opposite; they enjoyed it so much, and with such voracity, that the text did not survive its readers. To the contrary, a text may survive in huge numbers precisely because nobody read it, but simply were expected to own it, like the copy of Tolstoy’s War and Peace that sits on my shelf and gathers dust.

So how can we reconstruct a textual history with any accuracy if most of the texts are missing?

My research is very much twisted up in this problem, of representing that which is missing from the historical record. The material details of the manuscripts which do survive often provide the answers, even if they are only partial ones. Study of fragmented texts, damaged codices, and how they compare to similar, more complete books, can tell us about how such things were read, and used, and potentially how such books came to be lost.

The four surviving manuscripts of The Awntyrs off Arthure for example, reveal that three of them were produced as paper (rather than parchment) booklets. Two of these copies are scribal exemplars, meaning they were intended to be the copy-text for reproducing the poem. These have therefore not been read as actively as they may have been had they been copies owned and consistently read by a family, as is the case with the Thornton Awntyrs of the Lincoln Cathedral Library MS 91 (c.1430-50). This has evidence of significant wear and (quite literally) tear, with sections of the text missing. The only reason for its survival is perhaps its position in the very middle of Robert Thornton’s very large book, made up of 340 folios and 65 texts. Had the text circulated and remained as booklets, as I’m almost certain was the case, there is very little chance of these having survived.

Immateriality matters because of how this informs our reading of the material which does survive and vice-versa. How texts were used is important for thinking about what they meant in any given literary culture, and, as my research continues, I hope to be able to demonstrate that the survival of Awntyrs presents a window into a much more extensive literary culture, of texts being produced for active consumption by eager audiences of Arthurian romance, of a literary culture not necessarily represented by the vestigial manuscript record, but which can be reconstructed through a close examination of the texts and codices which remain.

This has been adapted from a blog post originally written for Material Witness, an interdisciplinary doctoral training programme for the interrogation of physical objects, spaces and environments by humanities researchers. You can view the original post here.