Manuscript canonicity

ABSTRACT

Some manuscripts containing Middle English possess canonicity in modern scholarship for their own sakes, that is, for their interest as objects rather than due to the canonicity of their contents. The combination of many surviving manuscripts and a historical position largely before the coming of moveable-type print creates this manuscript canonicity; in this respect, Middle English studies stand out from comparable fields. Manuscripts with their own canonicity tend to be atypically large, and tend to gather many works, works that are often themselves not canonical. Although manuscript canonicity only rarely affects literary canonicity, the growth of manuscript canonicity shows how canonicity of all kinds emerges from teaching and research, not works or manuscripts. If literary scholars turn their attention to objects other than works, such as datasets, canonicity will develop around the new objects as it has around manuscripts. While the study of less-examined codices is worthwhile for other reasons, it can only move the hierarchy of canonicity around, and it cannot remove the stratification. However, manuscripts also invite thought on the books from the period which are now lost – the majority of books. Because research cannot make them canonical, these absent codices might help us think around canonicity.

KEYWORDS:

• Middle English
• codicology
• manuscript studies
• Chaucer
• manuscript canonicity

The 2019 Norton Chaucer presents an attractive and accessible text of the poet’s works, building on E. Talbot Donaldson’s later twentieth-century student-directed editions. The cover features a particularly canonical and authoritative Chaucer: a close-up of his depiction in the early twentieth-century stained glass of Worcester Cathedral’s cloister. Stained glass inevitably suggests memorialisation and veneration, and the window probably was designed to drive home a particular idea of past literary authority, for the Chaucer light serves as one half of a pair portraying fourteenth-century writers. To Chaucer’s right, absent from the cover of the Norton but present in a parallel pose on the other side of the mullion at Worcester, stands John Wycliffe.¹ The conjunction of these two figures offered an important literary-historical touchstone to some strands of Victorian and Edwardian Christianity: they were, for instance, repeatedly painted, sometimes together, by Ford Madox Brown, whose Chaucer at the Court of Edward III adorns the cover of The Riverside Chaucer in its 2008 printing.² Chaucer is the only Middle English poet with unquestionable public canonicity, and some very particular ideas of him quietly persist on the surfaces of twenty-first-century editions.

Partway through the introduction of the Norton, on the thirtieth page, another striking image appears. Its caption reads ‘Opening of the Canterbury Tales, Ellesmere manuscript (ca. 1400), Huntington Library, MS El 26 C9.’³ The introduction, commendably, seeks to convey to the edition’s student audience the significance of this codex, which contains the best surviving text of the Canterbury Tales and holds an exalted place in Middle English studies.⁴ The manuscript image in The Norton Chaucer is not, however, the first written page of Ellesmere. It is a replica – a hand-drawn, imitative facsimile, not a photograph – of the opening of the Tales in a different manuscript.⁵ I do not point this out to denigrate a very helpful student edition.⁶ No doubt Norton will fix the problem in future copies. If, that is, there is a problem in the first place. Does it matter that the image is not the manuscript mentioned in its caption?

Of course, simply in order to model responsible referencing, an edition’s illustrations ought to be what they say they are, or ought to come as close as they can to being what they say they are. A small black-and-white printed photograph is not, in important ways, the large and colourful handcrafted page from which it was produced, but we all know to navigate this problem of representation in daily life, including in situations with more directly charged ethical stakes: recognising that news photographs of an atrocity are not the atrocity represented shouldn’t restrain our anger on learning of the atrocity, and so forth.

My question, though, aims a little deeper than this. Consider the intended audience of The Norton Chaucer: undergraduates, many of them encountering Middle English literature for the first time. Almost all of this audience lack the palaeographical skills necessary to read the hand in Ellesmere, which is, in any case, challengingly small in the printed reproduction. This audience also usually lacks the wider grasp of the fourteenth- and fifteenth-century codicological landscape necessary to see Ellesmere in context: such luxurious copies were, non-specialist readers should note, highly atypical, and audiences encountering the Tales in manuscript would normally have done so in smaller and often scruffier books. Someone can only truly appreciate Ellesmere’s import and impact if they have handled its peers.

What more would the right image of the right object tell undergraduate readers? The fakesimile image already conveys the idea that the Canterbury Tales survived initially in manuscripts, and already conveys a sense of what a fairly luxurious manuscript looks like. It looks like a handwritten object. It resembles the opening of the Tales in Ellesmere far more than it resembles the equivalent text in, for instance, the copy of Caxton’s first printed edition of the Tales in the library of Merton College, Oxford, or in the collapsing, worn-out copy of The Riverside Chaucer on my office bookshelf, or indeed in The Norton Chaucer itself.⁷ For undergraduate readers, this image will simply convey the idea that manuscripts are not print, unless we start training undergraduates in palaeography and codicology before we expose them to Chaucer. But at present the study of premodern English is shrinking – while I wrote this article, a major British university began to cut the area from its curriculum – and so the widespread introduction of palaeography into undergraduate pedagogy seems improbable.⁸ The wrong image does its job, therefore, and will likely continue to do its job.

The image in The Norton Chaucer represents a canonical manuscript. It conveys that manuscript’s importance, prestige, and authenticity. More than eighty known manuscript copies of some or all of The Canterbury Tales survive.⁹ Had the Norton included an image of any one of the other witnesses instead, reviewers probably would have wondered why Ellesmere did not appear. The very idea of Ellesmere, without the presence or accurate reproduction of Ellesmere, has developed a force of its own, much like the idea of Chaucer. Ellesmere has manuscript canonicity.

I set out here what manuscript canonicity is, and how it might work. Literary canonicity gives us some models for the workings of manuscript canonicity. I do not suggest that manuscript canonicity in turn drives literary canonicity. Rather, I trace how the growth of manuscript canonicity in recent decades might hint at the workings of canonicity of all kinds. Since it can attach itself to physical manuscripts as well as literary works, canonicity doesn’t ultimately inhere in literary works, or in objects. Instead, scholarship and teaching generate it. Because of their particular nature as a corpus, I propose that manuscripts containing Middle English show in an especially clear manner that canonicity emerges from the stratification of prestige in teaching and scholarship themselves. Premodern manuscripts also, however, offer us at least one tool that tempers canonicity: the epistemic humility prompted by contemplating the many manuscripts and works that don’t survive.

On Middle English studies

Something called Middle English studies serves as a case study here, not by default but for two good reasons. First, Middle English manuscripts drive this study because scholarship should render this field distinct, and avoid making it an unspoken centre for, or zone coterminous with, the medieval. Middle English studies does not appear here as an assumed norm; it appears precisely because it should not function as an assumed norm, but should rather be named, seen, and spoken of. This overt treatment assists Middle English studies itself, reminding us of its real objects and value: the world needs a good historical account of its most widely-learned language, if for nothing else than as a standpoint from which to see that that widely-learned status is far from neutral, natural, or eternal. This overt treatment also aids the other specialisms carried on under the headings of the medieval and medieval studies. By naming Middle English and saying why it is present, scholarship can both wring more insights out of it and avoid granting it a default status. Explicitly demarcating Middle English studies allows other fields to come into sharper focus. It also helps to avoid either anglocentrism or the type of reaction to anglocentrism that still remains determined and defined by England the place and by the language of English.

Additionally, and second, Middle English studies also serves as a case study here because it boasts a more developed and more revealing landscape of manuscript canonicity than most fields. As a result, examining Middle English helps us to see how canonicity can function when uncoupled from literary works. Diachronic comparison shows how Middle English studies is unusual in this regard. In Old English studies, there are not enough extant manuscripts to permit the growth of developed hierarchies, the ascent of some books above others in strata of prestige. Modern studies of Old English literature certainly display a hierarchy of canonicity among works: crudely, the poem we now call Beowulf sits at the top, above some lyric poems, with other poems somewhere in the middle and most prose texts at the bottom. Most Old English poems, in particular, and all the poems most popular today, survive in single copies. All specialists in early English know of the four relevant manuscripts and know their resonant nicknames: Junius, Exeter, Vercelli, Nowell. It is, however, the canonicity of works that creates the distinctions in status separating manuscripts that contain Old English. The Nowell Codex has prestige because it contains Beowulf.¹⁰ Since no other copy of Beowulf survives to be snubbed in favour of Nowell, this manuscript’s dignity rests wholly on its contents.

The arrival of moveable-type printing, first to English and shortly thereafter in England, roughly coincides with the changes usually taken to signal the end of Middle English. One can easily either over- or under-state the impact of print, which affected different spheres of activity at different rates.¹¹ Print certainly did, however, significantly change how evidence would go on to survive in the twenty-first century. Printed books are always one of a class of object, part of a print run, even if only one of that class happens to survive. Printed books, and early printed books especially, do display meaningful variation, which fine bibliographical scholarship often traces in copy-specific studies. However, the early printed books with the most varied print runs display less variation, across various aspects of text and layout, than the closest exemplar–copy and shared-exemplar pairs among manuscripts. The difference is so great that it constitutes a difference in kind.¹² By contrast, for most Middle English, manuscripts are normally all we have. And every manuscript really is unique: copy-specific study is the ground and starting-point for work with such objects. Middle English differs from the Old English period in the frequent survival of multiple copies of works: even relatively early works such as Poema Morale, Layamon’s Brut, Ancrene Wisse, and The Owl and the Nightingale survive in more than one manuscript. Middle English differs from subsequent Englishes because manuscript canonicity attaches to specific single objects (manuscripts), not to classes of object (print runs). The distinctive thing about Middle English studies, seen against these comparanda, is that Middle English largely precedes English moveable-type printing, yet some Middle English texts survive in many copies. In these conditions, manuscripts can develop canonicity of their own. Moreover, some can develop it not from the prestige of texts and authors, but through their physical, codexical features, and through the combinations of texts that they physically bring about. This phenomenon deserves a survey of some examples.

Some canonical manuscripts

Examples of manuscripts with canonicity can be broken down into three groups. First, where Middle English works survive in just one witness, canonicity continues to be tightly tied to the work, not the manuscript. The highly-canonical poems Sir Gawain and the Green Knight and Pearl survive today in one codex, together with the rather less canonical Patience and Cleanness (these are all retrospective, modern titles). The poems have given the book, London, British Library, MS Cotton Nero A.x, its standard names, the Pearl Manuscript or the Gawain Manuscript, depending on which poem a critic wishes to bring to the fore. It is never the ‘Patience Manuscript’ or the ‘Cleanness Manuscript’. In this case, textual canonicity and manuscript canonicity march together: we cannot know how things would stand if another copy of one or more of these poems survived. This canonicity resembles the canonicity of the Nowell Codex, derived from the poem we call Beowulf: merely delegated literary canonicity, not pure manuscript canonicity.

Now, the Pearl Manuscript does have real codicological interest: its programme of illustration is unusual, its hand strangely archaising.¹³ What’s more, Shamma Boyarin has made the intriguing suggestion that single-witness survival sometimes burnishes a work’s reputation for present-day appreciation by tacitly offering the uniqueness of an art object.¹⁴ A cynic might add the further thought that the literary work surviving in just one witness can only invite a limited amount of textual critical effort, even when that witness plainly errs, saving the reading critic from paranoia about exactly what they think they’re reading. In these two ways, perhaps single-manuscript survival can contribute to literary canonicity. That effect is not manuscript canonicity, however, and in the cases of Pearl, Sir Gawain, and many of their single-witness brethren, it is easy to identify elements within the text that have made them attractive to modern readers. The Pearl Manuscript possesses canonicity not through its own codicological merits, but by virtue of its contents.

Second, works which survive in more than one copy create space for manuscript canonicity to grow. Some codices possess a mixture of literary or textual canonicity and manuscript canonicity. Ellesmere occupies such a halfway-house position. Ellesmere has canonicity in part because it contains the most famous work by the period’s only famous poet, and because its text contains relatively fewer variants which, with textual-critical hats on, we might call errors and seek to eliminate in an edition which aimed to print a text closer to that composed by Chaucer. (This is by no means the only possible goal of editing, but it is one legitimate goal.) Yet Ellesmere is only one among a reasonable number of copies of that text. It is also the manuscript’s size, its beauty, and its unusual illustrative portraits of Chaucer’s pilgrims that have made it iconic.

Third and finally, beyond these mixed cases, there are examples of pure manuscript canonicity: cases in which manuscripts’ own physical features themselves propel them to scholarly fame and name-recognition. Indeed, naming is the surest sign of manuscript canonicity, especially when the name does not come from a literary work. The phenomenon of manuscript naming has received limited study, but Wilma Fitzgerald, writing primarily from an art-historical perspective, has observed that the numerical elements of most manuscript shelfmarks encourage it, and that frequency of use cements it. Fitzgerald used a Middle English example of the tendency of such names to compound and confuse, remarking that ‘the Cholmondeley-Delamere-Penrose Chaucer is in Tokyo’, a statement which has since, in turn, become inaccurate: the Cholmondeley-Delamere-Penrose-Takamiya Chaucer is now in New Haven, Connecticut.¹⁵

This category of pure manuscript canonicity merits further reflection, and manuscripts containing Middle English held in British libraries offer several more examples. Consider the naming of the Harley Manuscript, London, British Library, MS Harley 2253. As the number 2253 shows, there are plenty of Harley manuscripts, including many others which contain Middle English: the eighteenth-century collecting efforts of Robert and Edward Harley gathered more than 7,500 manuscript books which were purchased for the British Museum. In English studies, though, this particular codex is the Harley Manuscript. The interest of the texts in this book play a role in its fame, but it is the codexical fact that they are brought together by the book as a physical object that has garnered both it and them so much attention. In 1948, an edition of some of these poems could take The Harley Lyrics as its main title despite the fact that, as the editor duly acknowledges, ‘other Harleian manuscripts contain lyrics’.¹⁶ That concession rather understates the situation: an editor of John Donne or Ben Jonson might be surprised – perhaps in a salutary way – to hear that the really key lyrics in the Harleian collection are those in a fourteenth-century codex.

The Auchinleck Manuscript has acquired similar pure manuscript canonicity.¹⁷ Auchinleck contains a mixture of romances and hagiographies, copied in the first half of the fourteenth century. For that date, it stands out in its scale and in the tight, though not total, focus of its collection on narratives. Some of the works in Auchinleck are also found elsewhere. Moreover, and at least as significantly, many of them have not received high evaluations for literary quality. They are mostly romances, and even experts on them have sometimes framed them as culturally-revealing ‘pulp fictions’, invoking their lack of canonicity and their low status as justifications for their study.¹⁸ Yet, because it brought these low-prestige texts together in an unusually large physical mass and at an unusually early date in the surviving record, Auchinleck as an object certainly has manuscript canonicity. That canonicity has sometimes distorted editing. Editions of Sir Bevis of Hampton, for example, have tended to base themselves on Auchinleck’s text, since the Auchinleck copy of Bevis predates the alternatives and survives in a famous book. As Jennifer Fellows outlines in recent revisionary work, however, Auchinleck’s Bevis has no claim to greater textual authority, and in fact diverges significantly more from the Bevis tradition than some later texts do.¹⁹ For Bevis, a work so variable that almost every text of it offers a distinct version, Auchinleck counts as a witness as important as any other, but has no real claim to the definitive status it has achieved in past editing. For Bevis, manuscript canonicity has affected editorial work and has silently inflected the literary work as experienced by scholars and students. Manuscript canonicity doesn’t always impact literary canonicity, but in this case – for the relatively thin crowd who read Bevis – it did for many decades.

The Vernon Manuscript offers the biggest example of pure manuscript canonicity in Middle English studies.²⁰ It is mentioned in printed scholarship and online many times more often than the median manuscript containing Middle English. Vernon holds this position because it outclasses all other manuscripts containing Middle English in size.²¹ A few of the works it contains have literary canonicity. Except in cases in which Vernon serves as the only known witness to a work, however, textual criticism rarely values works in their specific Vernon texts. Vernon’s text of the A-version of the canonical poem Piers Plowman, for instance, hasn’t supplied the base text or copy text for any edition since the nineteenth century. The Vernon text of Piers Plowman also receives little dedicated study in textual-critical or literary-critical scholarship on the poem. The Yearbook of Langland Studies has published many articles focussed on particular texts of Piers, but has never carried any article specifically about the text of the poem in Vernon. Scholars have sensible textual-critical reasons to focus on other witnesses, and I do not suggest that they should do otherwise; I simply point out this pattern of scholarship to show that Vernon’s fame does not rest on the works it contains. As a witness to all of them at once, however, it becomes unignorably big, and fascinating for literary and cultural history. Vernon earned its reputation through brute size, and by gathering in so many works. In this case there is even a less-canonical shadow, the Simeon Manuscript, which garners much less attention for its own sake but travels in the train of its more famous cousin, like a codicological Troilus and Criseyde to Vernon’s Canterbury Tales.²²

As is the way with literary canonicity, so with manuscript canonicity: all three of these named examples, Harley, Auchinleck and Vernon, have atypical features. They have achieved fame precisely because they seem distinctive, and this atypicality might make them more valuable in themselves. Being so remarkable, though, they cannot tell us about normal reading or normal readers. Well-studied but unrepresentative, they stand out as beacons of manuscript canonicity in the wider and largely unlit prospect of many less-often discussed manuscripts containing Middle English.

An irony hangs around the persistence of manuscript canonicity: once upon a time, manuscripts were meant to save the field from the distortions of the canon. Several strands of scholarship from the 1980s onward implicitly or explicitly folded this into their agenda. One, drawing on Bernard Cerquiglini’s valorisation of the variation between different manuscript copies of works, was the ‘New Philology’.²³ Put crudely, this approach entailed a new focus on the individual manuscript as a package of cultural information equally valuable as any other, staying thoroughly disengaged from aesthetic or textual-critical questions that might value one manuscript over others or, at a more granular level, one variant over others. At roughly the same time, some textual-critical thought itself turned to look again, and with scepticism, at its conception of the idealised and unified work abstracted from surviving copies.²⁴ Meanwhile, the broader history of the book took off, with less attention to theory (indeed, in some though not all cases, perhaps partly as a flight from it). In premodern literary studies, the history of the book swept up with it the older, pre-existing fields of palaeography, codicology, and bibliography. Practitioners sought to flatten out literacy canonicity, declaring what Paul Eggert retrospectively describes as ‘a kind of book-democracy’.²⁵ This tendency runs through Ralph Hanna’s influential collection of papers, Pursuing History, for example, and becomes explicit at points in his introduction.²⁶ For Middle English, these trends cohered with an emerging line of scholarship in ‘manuscript studies’, traceable through the 1987 foundation of the Early Book Society, and the collections emerging from the York Manuscripts Conference in the 1980s.²⁷

These developments spurred on useful work and encouraged scholars to cultivate a more fine-grained sense of manuscript evidence. These developments did not, however, permit either an escape from or a flattening-out of canonicity. Instead, scholarship clustered around manuscripts just as it already had around works. Before the twentieth century’s second half, there had been nascent manuscript canonicity among palaeographers and textual critics; by the century’s end, strata of manuscript canonicity had formed in literary-critical work too. Research on obscure books, including obscure copies of canonical works, tends to find neglect; research on well-known books tends to find enthusiasm. Manuscript canonicity exists and persists, without mapping neatly onto, or following neatly from, literary canonicity.

Canonicity does not, then, inhere in literary works pure and simple: physical things can have canonicity too. If not in works, where does canonicity live? It lives in the logistical channels of research and in the reward structures of scholarship and teaching, not in their objects.

Canonicity’s practical bases

Scholarship has become very familiar with the assertion that this or that literary work has been left outside the canon; I observe this as someone who has made such assertions, and will do so again below. A persuasive strain of past arguments, however, point to teaching as the real driver of literary canonicity. John Guillory made this case in the 1990s, and Susan VanZanten Gallagher has used Guillory’s ideas to develop a model specifically shaped for pedagogical discussion.²⁸ Recently, Rachel Sagner Buurma and Laura Heffernan have come at these issues from a different angle, and yet drawn the consonant conclusion that the classroom plays a much larger role in shaping research and the canon than is normally acknowledged.²⁹ For literary canonicity, then, the pedagogical canon might be what really matters. Perhaps the pedagogical canon matters for manuscript canonicity too.

Manuscripts remain less central to teaching in Middle English studies than texts, but their role has grown over the past few decades. Most instructors might now be expected to show students some facsimile images of manuscript pages, and to ask students to think about the implications of manuscript transmission. A growing number of institutions have some kind of explicit history of the book teaching at the undergraduate level, which might at least touch on books before the age of print. Unsurprisingly, therefore, teaching also affects manuscript canonicity. Even more than research, teaching imposes selection. This is especially the case for general courses about literature from the period and for the long survey courses common in North American institutions. On either type of course, teachers might have to squeeze manuscripts in as visual aids rather than as a topic in their own right. Moreover, as the precariously-employed shoulder an ever-higher proportion of teaching, teaching contact time becomes pressured and preparation becomes more short term. The research–teaching link also weakens, not because precarious scholars produce inferior research, but because a lack of at least medium-term security makes publishing a greater struggle. What’s more, the physical existence of manuscripts in particular geographical places means that manuscript canonicity, in teaching even more than in research, raises an intensified set of issues about convenience and access.

Because manuscripts physically exist, scholars cannot shift the contours of manuscript canonicity in scholarship and teaching by buying editions of a new primary text. Important problems of access determine which manuscripts can appear in teaching and research, and accessibility and canonicity mutually reinforce one another in a spiralling relationship. Joan L. Brown has already briefly noted that accessibility affects literary canonicity.³⁰ It has far more influence on manuscript canonicity. Patterns of facsimile publication, online or in print, clearly distinguish famous books containing Middle English from ignored books. A scholar specialising in earlier centuries can survey a landscape of more comprehensive digitisation, and move on to the next question of what to do when (for example) all of a library’s early English manuscripts are available in facsimile, or when the four manuscripts containing most of the known verse in an early vernacular have been reproduced.³¹ A scholar specialising in later centuries will have patchy access to manuscripts, but might have surprisingly good access to early print in the form of the Early English Books Online (EEBO) project. EEBO, a collection of re-re-mediated black-and-white images digitised from microfilm, has its own history, foibles, and costs. The sense of total access to print that it offers is, finally, illusory, and its oddities have birthed debates in scholarship.³² It does, though, offer a kind of access to materials far closer to complete than the landscape of manuscripts containing Middle English. An early printed book from England with no digital surrogate is a pariah, while a manuscript containing Middle English with no digital surrogate is typical: the majority of manuscripts containing Middle English have not yet received facsimiles.

Furthermore, no clear, practical route towards such mass digitisation exists. Digital facsimiles come with costs. High-quality images, good metadata, and hosting all cost money. Hosting also uses energy and so contributes in some small way, so long as energy provision partially relies on fossil fuels, to the heating of the planet, just like efforts to ensure proper temperature and humidity for physical books.³³ Libraries must navigate a spectrum of options from detailed, multispectral, multi-angle technological investigations of individual codices – through slow, expensive, but high-quality digitisation with good metadata – to fast, cheap digitisation with hand photography and no metadata, sometimes carried out by researchers themselves.³⁴ All three of these approaches can bear fruit; all three have disadvantages.

In any case, any library with a meaningfully large collection of manuscripts cannot afford to digitise most of them: a library which has digitised every one of its premodern European manuscripts is almost guaranteed to have a small collection, in the tens or hundreds of manuscripts, and a large budget. Scholars themselves can win funding for purposive, occasional digitisation of items relevant to a successful grant application. Since, however, pre-existing research interests shape such digitisation, this approach doesn’t permit the serendipitous discovery that this or that book holds more interest than previously thought. Project-led digitisation tends to strengthen canonicity’s grip by making canonical manuscripts more available and leaving the less-canonical behind. We ought, too, to register the opportunity cost of digitisation, when training, conservation, and public engagement all also cry out for time and labour. I think some manuscripts make a strong case for digitisation, but it is a case, not an imperative: digitisation isn’t an automatic, unmixed moral good.

For some time to come, perhaps for a long time, most manuscripts containing Middle English will remain accessible only through personal visits to libraries. Both the chronic problem of climate change and the acute problem of the COVID-19 pandemic ought to make scholars think on the tenuousness of this method of access. It is already the case that a manuscript lacking a facsimile and located in an out-of-the-way library will receive less study. As travel is likely to become more difficult and more expensive in future, this practical aspect, the very much non-metaphorical landscape of scholarship, will become more important.

Meanwhile, I know of four print facsimiles of Ellesmere, plus the digital facsimile available online.³⁵ This codex has, therefore, five more facsimiles than the typical manuscript containing Middle English. Such a distribution of attention inevitably reinforces manuscript canonicity: if a researcher wants to conduct research from an institution without a local manuscript collection – that is, from nearly any institution – they will pragmatically lean on the manuscripts which can be accessed; if they want to show students a manuscript in a lecture or send them off to look at a manuscript, they will once again pragmatically lean on the manuscripts which can be accessed. The field thus proceeds through a self-reinforcing process in which manuscript canonicity prompts the making of facsimiles, which in turn strengthen existing patterns of manuscript canonicity. Digital facsimiles have significantly sped up this spiral process, but the spiral began as soon as physical facsimiles became possible.³⁶ The publication of diplomatic or near-diplomatic transcriptions has sometimes had a similar effect. In one volume, the late nineteenth-century Chaucer Society published six transcribed texts of the Canterbury Tales – and later supplemented this work with two more texts – to facilitate editing and in part from a desire to show how some available manuscripts differed from earlier editions. Their selection, however, elevated the status of the six and then eight witnesses chosen, with consequences that echoed down through twentieth-century textual criticism and codicology.³⁷

Middle English studies has also cultivated a subgenre of edited collections dedicated to particular manuscripts.³⁸ These create a subtler but significant type of accessibility. This is one of their great virtues: they make the particular manuscripts on which they focus more accessible by laying out and solving issues in the codicological evidence and by modelling how it might be discussed. Most simply but perhaps also most profoundly, they offer guides to the contents of manuscripts, which by mapping and isolating each contained text and by making decisions about where texts join and divide – by no means always an easy task – make the book legible for other scholarship.³⁹ The effect on further research resembles that of a facsimile. Like digitisation, the distribution of scholarly attention via such edited collections does, however, naturally follow pre-existing scholarly lines of interest, and so tends to reinforce manuscript canonicity.

An article on the topic of canonicity might conventionally now turn to discuss a non-canonical, un-digitised case study, valorising its genuine interest. Such a move would show the value of evidence passed over by canonicity. Oxford, Bodleian Library, MS Laud Misc. 486 would serve this purpose well, and for the spell of a few paragraphs I shall show how. This book has received few mentions in published scholarship and little focused discussion.⁴⁰ Before even opening this book, a scholar would note that, unusually and excitingly, it existed in something like its present size and exterior form by the second half of the fifteenth century. Most manuscripts containing Middle English have modern bindings, and indeed, in many cases, we cannot safely assume that the elements of what is now one unified book were unified by one binding within the period, or were even bound at all. MS Laud Misc. 486 itself, in fact, has a distinctly worn and dirtied first page, perhaps suggesting time spent ‘in use’ before any binding was applied.⁴¹ Now, though, it sports a fifteenth-century binding, of a style that rapidly fell out of use through the second half of the century.⁴² In this respect, it can already reveal more about what it was like to encounter Middle English verse in its time than Ellesmere, which has a post-medieval binding.

The book contains a copy of The Prick of Conscience, the most widespread English poem before print, and the Latin prose Liber regulae pastoralis (Pastoral Care) of Pope Gregory I. Both works were centrally important in the written culture in England in the fourteenth and fifteenth centuries, and both works are rarely read today. The same scribe copied both texts, in a hand modelled on anglicana formata and dating to the first half of the fifteenth century, probably the first quarter. The transition from The Prick of Conscience to the Pastoral Care occurs on conjoint leaves within the same quire – the same physical gathering of leaves – and so both parts of the book must come from the same production context, and were not simply bound together by chance. Yet the joining of the two texts was probably not the plan from the very beginning, for the quire in which the switch happens is shorter (eight leaves) than those before and after (twelve leaves, the standard length throughout most of the book).⁴³ The scribe’s choice to make up a shorter quire for the end of The Prick of Conscience looks like the act of someone finishing up, anxious not to waste parchment with an unduly long quire of mostly blank leaves, rather than the act of someone expecting to launch immediately into a new text. The manuscript can immediately reveal something, then, about the conjunction of these two texts, an event deliberate yet not planned from the very start.

No explicit evidence survives about who copied this book, or who caused it to be made; we do not know who its first owner was. Both texts are, though, in different ways, keenly engaged with Gregory’s topic, pastoral care. That shared concern might indicate that the first owner was a churchman with some level of pastoral responsibility. A later fifteenth-century note survives at the book’s end, however, to suggest that it was in a lay, familial context by that time: the note records the birth of ‘Margaret of Eston’ in 1334.⁴⁴ The manuscript well postdates 1334, of course, and the note therefore seems to preserve the birth record, copied from elsewhere, of someone by then long dead. Eston is in Yorkshire, but Robert E. Lewis and Angus McIntosh placed the language of this codex’s copy of The Prick of Conscience in Gloucestershire.⁴⁵ That’s a fair way for the book to travel. But it would be slightly odd for Margaret to be called ‘of Eston’ in the place itself: perhaps the inscription merely and rather frustratingly tells us that the book wasn’t in Eston.⁴⁶ It does, in any case, suggest the preservation of a family history. This manuscript probably had an initial reader-owner with pastoral responsibilities, but might then have found its way to lay readers and a family setting. The manuscript also contains evidence of a later reader aware of changes wrought in England’s church and liturgy. In the middle of the sixteenth century, an English version of the traditional Latin hymn ‘Veni creator spiritus’, probably translated by Thomas Cranmer, was folded into Cranmer’s new ordination ceremony for priests. A sixteenth-century hand copied the first quatrain of this translation into MS Laud Misc. 486.⁴⁷

The collocation of The Prick of Conscience and Gregory’s Pastoral Care in this manuscript also evidences, more so than many canonical manuscripts, the fifteenth century’s continuing interpenetration of Latin and English, with Latin rather than English as a normative written language. The Prick of Conscience itself contains frequent blocks of Latin quotation from authorities scattered throughout its English verse, occasionally quotes from others among Gregory’s works, and received, unusually for a Middle English work, an epitome-translation into Latin of its own.⁴⁸ As often, part of the value of seeing Middle English in this book lies in observing English in circumstances in which it was not a dominant or default tongue. This point has much value for both first- and second-language speakers of English today. Only Middle English can make this point, because much of the surviving evidence of Old English survives in a written form which diverged somewhat from speech and represented a kind of partial, incomplete standardisation, while from the later fifteenth century onward English once more developed tentative, incipient standardisation.⁴⁹ Only during the Middle English period, among all periods of the language’s history, was English fully socially displaced and orthographically decentralised. These valuable aspects of Middle English studies easily disappear if one consults only grand copies of The Canterbury Tales. While MS Laud Misc. 486 might seem a thoroughly nondescript manuscript from the first half of the fifteenth century, it really trails cultural tendrils stretching back to before the Black Death and forward to the Book of Common Prayer. It opens up stories about sociolinguistic history and the binding of books. It can tell us more about some of these matters than Ellesmere, Auchinleck, Vernon, or the Harley Manuscript can.

Such an argument, which could run longer and include more detail – this codex serves up rich evidence, and limitations on space have made me selective – would be worthwhile in its own way. It would not, however, offer a radical rejoinder to existing canonicity, for two reasons. First, The Prick of Conscience, like many other works that were widespread and successful in the period, jars against the tastes and priorities of the present-day academy. As Laura Saetveit Miles crisply remarks, kicking off an incisive discussion of the potential canonical placement of Middle English devotional prose, ‘Most medievalists would agree that the modern canon of medieval literature does not perfectly align with the medieval canon of medieval literature – and nor should it’.⁵⁰ There may even be a direct opposition between past and present popularity, persistent across linguistic traditions and continents.⁵¹ The listing of the actual landscape of Middle English textual success with an aside on this mismatch has become a recurring topos in past studies.⁵² The three most successful works in Middle English are, ordered by decreasing numbers of surviving copies, the Wycliffite Bible – the canon, and probably the longest Middle English work – the Middle English prose Brut – a part-mythical history of Britain and then England – and The Prick of Conscience. All three offer much to studies of fourteenth- and fifteenth-century culture, and all three urgently need more research into their complex, multifarious textual and material forms. Automatically rebuilding a curriculum around them would, though, reinscribe some orders and priorities from the period itself. The same thing would happen if we turned to what intellectuals in the period spent most of their time reading: the great mass of manuscripts containing writings in Latin.⁵³ Cultural history might have an obligation to reflect the interests of the time it studies. The history of the book must attend to historical patterns of production and distribution. But Middle English studies has, woven together with its commitments to those fields, literary-critical fealties. Most practitioners will in the final analysis admit to choosing texts for teaching in a negotiated process which balances at least all of the following factors: historical significance, student interest, pedagogical value, thematic thrust, and literary qualities.

Second, there is the simpler but more profound problem that a revised selection remains a choice. We could build a field of a rather different sort around objects such as MS Laud Misc. 486. But that field would still have its omissions. Choice forges canonicity. Canonicity might, therefore, never really disappear so long as formal education exists. For students to receive a grounding in a subject within a sensible time limit, scholars must choose works from a much larger range. To communicate research to other researchers, scholars must choose examples from their broader studies. These choices become the canon for their audiences. A certain sort of canonicity is, then, a necessary byproduct of learning.

Unchosen and unchoosable

The field’s pedagogy and its institutions produce canonicity, and that canonicity will reattach itself to whatever objects are chosen for study. It is all too easy to slip into thinking that literary works themselves straightforwardly possess canonicity, and that scholarship can therefore escape canonicity through its choice among works. This idea itself rests on a reified, atomised conception of the work of the sort associated with the New Criticism. In fact canonicity emerges from choice, and choice becomes necessary due to teaching. If literary scholarship turns to new works, or to things other than works, that turn will not remove the problem of the canon: canonicity will simply attach itself to literary critics’ newly-chosen objects. I like manuscripts, and I see good reasons to study them, but they will not save literary studies from itself, or, for that matter, from the priorities of governments and funding bodies.

Complaints about canonicity might sound like the set-up for an argument in favour of quantification and abstraction. Such quantification in work with manuscripts is now a long-standing practice and one that predates widespread computing. It certainly yields useful results.⁵⁴ However, it probably produces canonical corpora. Again, the canonicity remains, simply attached to other objects. Similarly, the rise of the digital humanities means that particular literary datasets are probably developing canonicity at present. By noting this, I don’t mean to suggest that research should avoid new methods or new objects: new methods and new objects might have other virtues. I simply caution that we should not expect such new methods or new objects to solve this particular problem.

At this point, one might understandably despair. As a first, tempering response, we might note that this is at least a picture of canonicity which somewhat absolves scholars themselves. Scholars in general – and precarious scholars especially – have little say in canonicity, which is pressed upon us by the structures within which we teach and seek to further our careers. But there is at least one other, more positive route for response. Through the contemplation of manuscripts, scholars have access to one particular counterweight to canonicity which resists the instrumentalism of pedagogical or research-based imperatives: the missing books, the vast number of manuscripts which are lost.⁵⁵

These missing books have great importance, for the most truly normal manuscript containing Middle English is fully absent.⁵⁶ ‘Fully’ here means not even surviving as a fragment: even manuscripts surviving as one fragmentary leaf are extraordinary and, in their continuing existence, unrepresentative. Time and chance have already chosen every extant manuscript, and have already made the most powerful distinction of canonicity possible: the distinction between the surviving and the absent. We cannot, in important and obvious ways, ever choose the missing manuscripts for teaching or for research. In many ways, they therefore constitute a stumbling block to our understanding. In dealing with canonicity, though, this inaccessibility becomes helpful. Since we cannot select them for study or teaching, we can hold them in mind and point them out as we choose and discuss what does survive. The missing manuscripts might inculcate epistemological humility or, to put it more pithily, keep our thinking honest. The problem they present might also prompt fruitful thought about other periods: compared to the uncomfortably rich trail of data generated by people living today, even the later twentieth century looks like the victim of large-scale information loss. In this respect, contrasted with present-day data wealth – a firmly non-metaphorical wealth, hoarded by particular companies – all centuries before the present perhaps sit closer to the state of premodernity than we habitually think. On this point, as is often the case, thought about manuscripts containing Middle English brings a wider truth into focus.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 The two lights above Chaucer and Wycliffe depict two fourteenth-century bishops of Worcester, Thomas Cobham (d. 1327, donor of Oxford’s first university library), and Henry Wakefield (d. 1395); the four windows together present a striking, if implicit, local ecclesiastical sense of historical priorities. ‘William Longland’ appears in a smaller, lower light to Chaucer’s left, with his eyes raised towards Chaucer. I am very grateful to David Morrison of Worcester Cathedral, who kindly corresponded with me about the glass and provided a photograph of the upper lights. For advice on the article as a whole, I would like to thank the editors and the anonymous reviewer.

2 The Riverside Chaucer, gen. ed. Larry D. Benson (Boston, MA: Houghton Mifflin, 1987; repr. Oxford: Oxford University Press, 2008), cover (Ford Madox Brown, Chaucer at the Court of Edward III, 1847–51, oil on canvas, 372 × 296 cm, Art Gallery of New South Wales). See Velma Bourgeois Richmond, ‘Ford Madox Brown’s Protestant Medievalism: Chaucer and Wycliffe’, Christianity and Literature, 54(2005), pp. 363–96, doi:10.1177/014833310505400308. The same painting also adorns the cover of Susanna Fein and David Raybin (eds), Chaucer: Contemporary Approaches (University Park, PA: Penn State University Press, 2010).

3 The Norton Chaucer, gen. ed. David Lawton (New York, NY: Norton, 2019), p. 30.

4 The Huntington Library displays Ellesmere in its permanent, public exhibition, and hosts a publicly available facsimile of it online: <https://hdl.huntington.org/digital/collection/p15150coll7/id/2367> [Date accessed 9 February 2021]; the opening of the Tales can be found at MS El 26 C9, f. 1^r. When I write ‘best’, I do not mean that the Ellesmere text is free from error.

5 The page imitated is London, British Library, MS Harley 7334, f. 1^r, which can be consulted in digital facsimile at <http://www.bl.uk/manuscripts/Viewer.aspx?ref=harley_ms_7334_f001r> [Date accessed 9 February 2021].

6 See, for example, David Raybin, review of ‘David Lawton (ed.), The Norton Chaucer’, Studies in the Age of Chaucer, 42 (2020), pp. 421–5, doi:10.1353/sac.2020.0026.

7 Oxford, Merton College, early printed book 111.C.9; Riverside Chaucer, p. 23. On canonicity in the presentation of the start of the Tales in Riverside, see Stephanie Trigg, ‘Opening the Canterbury Tales: Form and Formalism in the General Prologue’, in Thomas A. Prendergast and Jessica Rosenfeld (eds), Chaucer and the Subversion of Form (Cambridge: Cambridge University Press, 2018), pp. 182–99, doi:10.1017/9781108147682.010. See also Daniel Davies, ‘The Social Life of the Riverside Chaucer’, Avidly (2021) <http://avidly.lareviewofbooks.org/2021/06/23/the-social-life-of-the-riverside-chaucer> [Date accessed 24 June 2021].

8 Anna McKie, ‘From Richard III to Boycott: Where did it Go Wrong at Leicester?’, Times Higher Education, 1 July 2021 <https://www.timeshighereducation.com/news/richard-iii-boycott-where-did-it-go-wrong-leicester> [Date accessed 20 September 2021].

9 Michael G. Sargent, ‘What Do the Numbers Mean? Observations on Some Patterns of Middle English Manuscript Transmission’, in Margaret Connolly and Linne R. Mooney (eds), Design and Distribution of Late Medieval Manuscripts in England (Woodbridge: Boydell and Brewer, 2008), pp. 205–44 (pp. 206, 210).

10 London, British Library, MS Cotton Vitellius A.xv.

11 On the now-long history of debates over print’s effects, see Sabrina Alcorn Baron, Eric N. Lindquist, and Eleanor F. Shevlin (eds), Agent of Change: Print Culture Studies after Elizabeth L. Eisenstein (Amherst, MA: University of Massachusetts Press, 2007).

12 Tim William Machan, Textual Criticism and Middle English Texts (Charlottesville, VA: University Press of Virginia, 1994), pp. 139–40.

13 For work on the illustrations, with references to past studies, see Maidie Hilmo, ‘Did the Scribe Draw the Miniatures in British Library, MS Cotton Nero A.x (the Pearl-Gawain Manuscript)?’, Journal of the Early Book Society, 20(2017), pp. 111–36. On the hand, see Jane Roberts, ‘The Hand and Script’, in The Cotton Nero A.x Project (2010) <http://people.ucalgary.ca/~scriptor/cotton/index.html> [accessed 17 June 2021].

14 Shamma Boyarin, ‘The Alexander Romance in the Age of Scribal Reproduction: The Aesthetics and Precariousness of a Popular Text’, in Heather Blurton and Dwight F. Reynolds (eds), Bestsellers and Masterpieces: The Changing Medieval Canon (Manchester: Manchester University Press, 2022), pp. 199–214 (pp. 208–9).

15 Wilma Fitzgerald, Ocelli Nominum: Names and Shelf Marks of Famous/Familiar Manuscripts (Toronto: Pontifical Institute of Mediaeval Studies, 1992), p. 1; New Haven, Beinecke Library, MS Takamiya 32. The counterintuitive disyllabic pronunciation of Cholmondeley would make this name for the book especially unhelpful in oral discussion.

16 The Harley Lyrics, ed. G. L. Brook (Manchester: Manchester University Press, 1948), p. 1. For a more modern edition of this manuscript’s complete contents, which contextualises the famous Middle English poems with their now less well-known Latin and French fellow-texts, see The Complete Harley 2253 Manuscript, ed. Susanna Greer Fein, trans. David Raybin and Jan Ziolkowski, 3 vols (Kalamazoo, MI: Medieval Institute Publications, 2015). The title of this more recent edition perhaps registers the greater focus on manuscripts in present-day scholarship.

17 Edinburgh, National Library of Scotland, MS Advocates 19.2.1.

18 Nicola McDonald (ed.), Pulp Fictions of Medieval England: Essays in Popular Romance (Manchester: Manchester University Press, 2004).

19 Bevis of Hampton, ed. Jennifer Fellows (Oxford: Oxford University Press, 2017), lxvii–lxxvii; see also Jennifer Fellows, ‘Author, Author, Author  …  An Apology for Parallel Texts’, in Vincent McCarren and Douglas Moffat (eds), A Guide to Editing Middle English (Ann Arbor, MI: University of Michigan Press, 1998), pp. 15–24.

20 Oxford, Bodleian Library, MS English poetry a. 1.

21 The Vernon Manuscript’s leaves measure roughly 545 × 395 millimetres, and it is thought originally to have contained more than 400 of these. About seventy leaves are now lost, but even in its present-day form the codex weighs twenty-two kilograms.

22 London, British Library, MS Add. 22283.

23 See Stephen G. Nichols, ‘Introduction: Philology in a Manuscript Culture’, Speculum, 65 (1990), pp. 1–10, and the other essays in the same issue; Bernard Cerquiglini, Éloge de la variante: Histoire critique de la philologie (Paris: Éditions du Seuil, 1989).

24 Machan, Textual Criticism, pp. 136–76 and throughout.

25 Paul Eggert, ‘Brought to Book: Bibliography, Book History and the Study of Literature’, The Library, 7th series, 13 (2012), pp. 3–32 (p. 28), doi:10.1093/library/13.1.3.

26 Ralph Hanna, Pursuing History: Middle English Manuscripts and Their Texts (Stanford, CA: Stanford University Press, 1996), e.g. p. 3.

27 The first proceedings volume from these conferences is Derek Pearsall (ed.), Manuscripts and Readers in Fifteenth-Century England: The Literary Implications of Manuscript Study (Cambridge: D. S. Brewer, 1983); such volumes eventually formed a series, starting with A. J. Minnis (ed.), Latin and Vernacular: Studies in Late-Medieval Texts and Manuscripts (Cambridge: D. S. Brewer, 1989).

28 John Guillory, Cultural Capital: The Problem of Literary Canon Formation (Chicago, IL: University of Chicago Press, 1993); Susan VanZanten Gallagher, ‘Contingencies and Intersections: The Formation of Pedagogical Canons’, Pedagogy, 1 (2001), pp. 53–67, doi:10.1215/15314200-1-1-53.

29 Rachel Sagner Buurma and Laura Heffernan, The Teaching Archive: A New History for Literary Study (Chicago, IL: University of Chicago Press, 2021), pp. 210–13, doi:10.7208/chicago/9780226736273.001.0001.

30 Joan L. Brown, ‘Constructing our Pedagogical Canons’, Pedagogy, 10 (2010), pp. 535–53 (pp. 541–2), doi:10.1215/15314200-2010-006.

31 Elaine Treharne, ‘Raw Materials: The Role of Paleography in Medieval Studies’, in Stefan Jurasinski and Andrew Rabin (eds), Languages of the Law in Early Medieval England: Essays in Memory of Lisi Oliver (Leuven: Peeters, 2019), pp. 155–75, doi:10.2307/j.ctv1q26xqz.17.

32 Ian Gadd, ‘The Use and Misuse of Early English Books Online’, Literature Compass, 6 (2009), pp. 680–92, doi:10.1111/j.1741-4113.2009.00632.x; Michael Gavin, ‘How to Think about EEBO’, Textual Cultures, 11 (2017), pp. 70–105, doi:10.14434/textual.v11i1-2.23570; Peter C. Herman, ‘EEBO and Me: An Autobiographical Response to Michael Gavin, “How to Think about EEBO”’, Textual Cultures, 13 (2020), pp. 207–16, doi:10.14434/textual.v13i1.30078.

33 Joshua Calhoun, The Nature of the Page: Poetry, Papermaking, and the Ecology of Texts in Renaissance England (Philadelphia: University of Pennsylvania Press, 2020), pp. 140–5.

34 Andrew Prescott, ‘Slow Digitisation and the Battle of the Books’, in Catherine E. Karkov (ed.), Slow Scholarship: Medieval Research and the Neoliberal University (Cambridge: D. S. Brewer, 2019), pp. 143–61, doi:10.1017/9781787447042.008; Daniel Wakelin, ‘A New Age of Photography: “DIY Digitization” in Manuscript Studies’, Anglia, 139 (2021), pp. 71–93, doi:10.1515/ang-2021-0005.

35 The Ellesmere Chaucer Reproduced in Facsimile, 2 vols (Manchester: Manchester University Press, 1911); The Ellesmere Manuscript of Chaucer’s ‘Canterbury Tales’: A Working Facsimile, ed. Ralph Hanna (Woodbridge: D. S. Brewer, 1989); ‘The Canterbury Tales’: The New Ellesmere Chaucer Monochromatic Facsimile of Huntington Library MS EL 26 C 9, ed. Martin Stevens and Daniel Woodward (San Marino, CA: Huntington Library, 1997). A colour facsimile for collectors was produced in Tokyo as part of the same project: ‘The Canterbury Tales’: The New Ellesmere Chaucer Facsimile (Tokyo: Maruzen-Yushodo, 1995). The Huntington Library’s digital facsimile can be found at <https://hdl.huntington.org/digital/collection/p15150coll7/id/2359> [Date accessed 25 May 2021].

36 Treharne, ‘Raw Materials’, p. 165.

37 H. L. Spencer, ‘F. J. Furnivall’s Six of the Best: The Six-Text Canterbury Tales and the Chaucer Society’, Review of English Studies, n.s., 66 (2015), pp. 601–23, doi:10.1093/res/hgv004.

38 Susanna Fein, in particular, has coordinated several examples of this type of publication, for which scholarship owes her a significant debt of gratitude. Instances include Susanna Fein (ed.), Studies in the Harley Manuscript: The Scribes, Contents, and Social Contexts of British Library MS Harley 2253 (Kalamazoo, MI: Medieval Institute Publications, 2000); Susanna Fein and Michael Johnston (eds), Robert Thornton and His Books: Essays on the Lincoln and London Thornton Manuscripts (Woodbridge: Boydell and Brewer, 2014); Susanna Fein (ed.), The Auchinleck Manuscript: New Perspectives (Woodbridge: Boydell and Brewer, 2016); Susanna Fein (ed.), Interpreting MS Digby 86: A Trilingual Book from Thirteenth-Century Worcestershire (Woodbridge: Boydell and Brewer, 2019). The Vernon Manuscript has prompted not one but two collections of this sort: Derek Pearsall (ed.), Studies in the Vernon Manuscript (Cambridge: Brewer, 1990); and Wendy Scase (ed.), The Making of the Vernon Manuscript: The Production and Contexts of Oxford, Bodleian Library, MS Eng. poet. a. 1 (Turnhout: Brepols, 2013).

39 For example, Susanna Fein, ‘The Contents of Robert Thornton’s Manuscripts’, in Susanna Fein and Michael Johnston (eds), Robert Thornton and His Books: Essays on the Lincoln and London Thornton Manuscripts (Woodbridge: Boydell and Brewer, 2014), pp. 13–65.

40 The manuscript is one of the copies adduced in Angus McIntosh, ‘Two Unnoticed Interpolations in Four Manuscripts of the Prick of Conscience’, Neuphilologische Mitteilungen, 77 (1976), pp. 63–78. One researcher has used its rhyming opening summary of The Prick of Conscience as evidence of the mnemonic absorption of that poem: Daniel Sawyer, Reading English Verse in Manuscript c.1350–c.1500 (Oxford: Oxford University Press, 2020), pp. 56–7. The book is also notable for its striking ratio of weight given its size: see Sawyer, Reading English Verse in Manuscript, pp. 102–3.

41 Specifically, MS Laud Misc. 486, f. 1^r is notably dirtied compared to its conjoint counterpart, the other half of the first quire’s outer bifolium, f. 10^v.

42 Oak boards (Szirmai type [b]) covered in leather (corner turn-ins are Szirmai type [g]), with white leather sewing supports and endband supports (pathing Szirmai type [b]), inner board lacing paths mix (Szirmai types [E] and [B]). For these typologies, see J. A. Szirmai, The Archaeology of Medieval Bookbinding (Farnham: Ashgate, 1999), pp. 219, 223, 231. On the shift in binding style which helps to date this binding, see Nicholas Hadgraft, ‘English Fifteenth-Century Bookbinding Structures’, unpublished PhD thesis (University College, London, 1998), p. 245; for its possible cause, see Alexandra Gillespie, ‘Bookbinding’, in Alexandra Gillespie and Daniel Wakelin (eds), The Production of Books in England 1350–1500 (Cambridge: Cambridge University Press, 2011), pp. 150–72 (p. 170), doi:10.1017/cbo9780511976193.009.

43 The manuscript’s collation is I¹⁰, II–X¹², XI⁸, XII–XIV¹², XV¹⁰, XVI perhaps¹⁰ (perhaps -3-5, -8-10). The transition between texts occurs at ff. 122^r–123^r, that is, in the middle of quire XI rather than at a quire boundary. This is, however, the one atypical quire of eight leaves in a long run of quires of twelve leaves.

44 MS Laud Misc. 486, f. 175^v: ‘Margareta de eston nata erat in die sancte mathie apostoli inter horam x.am et xj.am anno regis Eddwardi iijii post conquestum viij’ (‘Margaret of Eston was born on the day of St Matthew the Apostle, between the tenth and eleventh hours, in the eighth year of the reign of Edward III’), in a mixed hand of the second half of the fifteenth century. Since the new year began on 25 March, to a fifteenth-century person Edward III would have begun his reign in January 1326 (our January 1327), and so this date is 21 September 1334.

45 Robert E. Lewis and Angus McIntosh, A Descriptive Guide to the Manuscripts of the ‘Prick of Conscience’, Medium Ævum Monographs, New Series, 12 (Oxford: Society for the Study of Medieval Languages and Literature, 1982), p. 106. Scribes could, of course, move, potentially taking their idio(grapho)lect with them, so dialectal localisation of this sort must be treated with some care.

46 On this point I differ from Lewis and McIntosh, Descriptive Guide, p. 106.

47 MS Laud Misc. 486, f. 15^r, ‘Com holy goste eternall god proceedynge from above … ’. Diarmaid MacCulloch, Thomas Cranmer: A Life, rev. edn (New Haven, CT: Yale University Press, 2016), p. 331. Cranmer first published this translation in his 1550 ordinal – The forme and maner of makyng and consecratyng of Archebishoppes, Bishoppes, Priestes and Deacons (London: Richard Grafton, 1549), STC2 16462 – and it was incorporated in subsequent versions of the Book of Common Prayer. On the complexities of confessional affiliation arising from sixteenth-century uses of earlier manuscripts, see Margaret Connolly, Sixteenth-Century Readers, Fifteenth-Century Books (Cambridge: Cambridge University Press, 2019).

48 Ralph Hanna and Sarah Wood (eds), Richard Morris’s ‘Prick of Conscience’: A Corrected and Amplified Reading Text, Early English Text Society, Original Series, 342 (Oxford: Oxford University Press, 2013), p. 407; Lewis and McIntosh, Descriptive Guide, pp. 163–8.

49 Merja Stenroos and Jeremy J. Smith, ‘Changing Functions: English Spelling before 1600’, in Vivian Cook and Des Ryan (eds), The Routledge Handbook of the English Writing System (London: Routledge, 2017), pp. 146–63 (pp. 151, 161–2); Mark Faulkner, ‘Quantifying the Consistency of “Standard” Old English Spelling’, Transactions of the Philological Society, 118 (2020), 192–205, doi:10.1111/1467-968X.12182.

50 Laura Saetveit Miles, ‘Canon, Anon., or a Nun? Queering the Canon with Medieval Devotional Prose’, Studies in the Age of Chaucer, 42 (2020), pp. 295–310 (p. 295), doi:10.1353/sac.2020.0009.

51 Heather Blurton and Dwight F. Reynolds, ‘Introduction’, in Heather Blurton and Dwight F. Reynolds (eds), Bestsellers and Masterpieces: The Changing Medieval Canon (Manchester: Manchester University Press, 2022), pp. 1–20 (p. 1).

52 Miles, ‘Canon, Anon., or a Nun’, pp. 295–6; Thomas A. Prendergast, ‘Canon Formation’, in Marion Turner (ed.), A Handbook of Middle English Studies (Oxford: Wiley, 2013), pp. 239–50 (pp. 239–40), doi:10.1002/9781118328736.ch15; Sargent, ‘What Do the Numbers Mean?’

53 A soup to nuts – undergraduate to doctorate – literary (as opposed to historical) discipline designed around Latin writings of this period would also be impractical given the competencies and training opportunities (un)available in the present-day academy and (at least in the UK) the gross inequalities which have persisted and indeed intensified in language learning in primary and secondary education: see further Mark Thakkar, ‘Duces caecorum: On Two Recent Translations of Wyclif’, Vivarium, 58 (2020), pp. 357–83 (pp. 382–3), doi:10.1163/15685349-12341391.

54 J. P. Gumbert, ‘Fifty Years of Codicology’, Archiv für Diplomatik, 50 (2004), pp. 505–526, doi:10.7788/afd.2004.50.jg.505.

55 The number of manuscripts surviving but as yet undiscovered is comparatively tiny, but is probably greater than zero; a recent example is Edinburgh, Advocates Library, Abbotsford MS, which returned over the scholarly horizon in 2004: Osbern Bokenham, Lives of Saints, ed. Simon Horobin, 3 prospective vols, vol i, Early English Text Society Original Series 356 (2020), pp. xiii, xxii–xxvi.

56 Scholars disagree in their estimates of loss rates for European manuscripts in this period, and disagree over the best methods of estimation, but nevertheless agree that the missing books significantly outnumber those that survive. See, for example, Uwe Neddermeyer, Von der Handschrift zum gedruckten Buch: Schriftlichkeit und Leseinteresse im Mittelalter und in der frühen Neuzeit; quantitative und qualitative Aspekte (Wiesbaden: Harrassowitz, 1998); David D’Avray, ‘Printing, Mass Communication, and Religious Reformation: The Middle Ages and After’, in Alexandra Walsham and Julia Crick (eds), The Uses of Script and Print, 1300–1700 (Cambridge: Cambridge University Press, 2004), pp. 50–70; Eltjo Buringh, Medieval Manuscript Production in the Latin West: Explorations with a Global Database (Leiden: Brill, 2011); and John L. Cisne, ‘How Science Survived: Medieval Manuscripts’ “Demography” and Classic Texts’ Extinction’, Science, 307.5713 (2005), pp. 1305–7, doi:10.1126/science.1104718, and critiques: Nicholas D. Pyenson and Lewis Pyenson, ‘Treating Medieval Manuscripts as Fossils’, Science, 309.5735 (2005), pp. 698–701, doi:10.1126/science.309.5735.698e, and Georges Declercq, ‘Comment on “How Science Survived: Medieval Manuscripts’ ‘Demography’ and Classic Texts’ Extinction”’, Science, 310.5734 (2005), p. 1618, doi: 10.1126/science.1117462. Estimates of loss rates for works, rather than or alongside loss rates for copies, are much rarer, but some can be found for chivalric and heroic narratives in Mike Kestemont and others, ‘Forgotten Books: The Application of Unseen Species Models to the Survival of Culture’, Science, 375.6582 (2022), pp. 765–9, doi:10.1126/science.abl7655.