Abstract
This article revisits the earliest manuscripts of the ‘History of Recent Events in England’ (Historia novorum in Anglia) written by Eadmer of Canterbury during the first decades of the twelfth century. Today, it is widely recognised as one of the most important pieces of historical writing produced in England during the period following the Norman Conquest. The article is the first in over twenty years to offer an in-depth codicological study of the Historia novorum’s principal manuscripts, and it provides a critical corrective to the canonical studies of Richard Southern (1963/90), whose arguments still form the communis opinio in present-day scholarship. Analysis shows that, contrary to established scholarly consensus, Eadmer did not compose his work in two distinct versions separated by a five-year writing hiatus, but that he kept writing continuously and produced one or more interim redactions of the Historia novorum that continued the narrative up to key moments in the history of early twelfth-century England.
I believe that a great favour has been rendered to posterity by those who committed to the memory of the written word a record of the events of their own time for the benefit of those eager to learn [about them] in the future … With this in mind, I have decided briefly to commemorate by the service of the pen the things I have heard and seen with my own eyes.1
These words were written during the first quarter of the twelfth century by Eadmer, monk-historian and cantor of Christ Church cathedral priory, Canterbury, whose archbishops at that time, and until the end of Eadmer’s life, were competing with the prelates of Y ork over the episcopal primacy in England (primas totius regni Anglorum) .2 Relying on the service of his pen (stili officium), Eadmer produced various hagio- graphical and historiographical writings, the most prominent of which are his ‘Life of Anselm’ (Vita Anselmi) and the ‘History of Recent Events in England’ (Historia novorum in Anglia, hereafter HNov) that both revolve around Anselm, Canterbury’s celebrated archbishop and a former prior and abbot of Le Bec in Normandy.3
Eadmer and his writings have received much scholarly attention, resulting in a large number of studies, editions and translations of most of his known works, including the Vita Anselmi.4 The HNov has always been a fundamental source for scholars and students of post-Conquest England, being celebrated widely by modern historians as ‘the first piece of large-scale historical writing in England after Bede’.5 Indeed, even some of Eadmer’s twelfth-century contemporaries looked upon it with a sense of admiration. For example, William of Malmesbury in his Gesta regum Anglorum praised the HNov as ‘a work composed arduously over the course of many a night, sober and erudite in style, … in which Eadmer presented students with a full and valuable history of events down to the death of Archbishop Anselm’.6 According to William, England by the turn of the twelfth century had failed to produce a reliable historian–by which he meant someone writing in Latin–since the Venerable Bede, except for Eadmer. The history of events separating Bede’s account from when Eadmer picked up his pen several centuries later, much to William’s disapproval, was neglected and left to ‘limp along without the patronage of the written word’.7
Despite the HNov’s pride of place within the historiographical production of post-Conquest England, the standard edition remains that published by Martin Rule in 1884, whose rendition of Eadmer’s ‘sober’ Latin has been translated into various modern vernaculars, including English, French and, most recently, Italian.8 Whilst Rule’s edition continues to provide the principal source-text for scholarship on Eadmer and his HNov, there have been few attempts during recent decades to re-examine the surviving manuscript evidence as a way of understanding better the text itself and its history of composition and dissemination—an important approach that is both distinct from and complementary to the many excellent palaeographical studies on scribal identification and scribal practice at Christ Church, Canterbury during the eleventh and twelfth centuries.9 The present article is the first in over two decades to return ad fontes and revisit the earliest manuscripts first-hand, and the first in over a century to offer a codicological study of all the known autographs, fair copies, fragments and so-called working copies of the HNov produced before the end of the twelfth century.
The HNov has generally been assumed to have enjoyed limited circulation during the twelfth century, its manuscript tradition and dissemination having been confined primarily—if not exclusively—to Eadmer’s home monastery of Christ Church, Canterbury. It is here, in the monks’ library, that the HNov might have been studied in situ by William of Malmesbury at some point before 1125, though this version of events is perhaps less likely than sometimes assumed.10 Whatever the extent of the HNov’s reception and impact during the decades following its ‘publication’,11 the extant manuscript evidence is modest by any standard.12 Only two complete textual witnesses are known to survive from the twelfth century, both of which were collated by Rule to establish the HNov’s text for his edition in 1884.13 The first is Cambridge, Corpus Christi College, MS 452 (hereafter MS CCCC 452), now widely accepted as Eadmer’s autograph, though parts of it were penned by Canterbury scribes other than Eadmer.14 The second is London, British Library, MS Cotton Titus A. ix (hereafter MS BL Titus A. ix), a manuscript dated to the later twelfth century.15 In addition, we have a single-sheet fragment of the HNov copied by one of the scribes involved in the production of MS CCCC 452—possibly Eadmer himself–which today survives on the flyleaf of Cambridge, Corpus Christi College, MS 341 (hereafter MS CCCC 341).16 Fourth and finally, there is a (partial) autograph of Eadmer’s second major work, the Vita Anselmi, copied alongside a selection of his so-called ‘minor works’ in Cambridge, Corpus Christi College, MS 371 (hereafter MS CCCC 371).
Most of these manuscripts have been examined in some detail over the course of the last one and a half centuries. In 1886, Rule swiftly followed his edition of the HNov with a book-length article on the text’s gradual revision and expansion during the final years of Eadmer’s life, for which he returned directly to the manuscript evidence.17 Three quarters of a century later, Richard Southern published a major reassessment of Eadmer’s manuscripts. In the appendices of his seminal monograph Saint Anselm and His Biographer (1963), Southern disagreed vehemently with the findings of his nineteenth-century predecessor, and he also cautiously identified Eadmer’s handwriting.18 Southern’s response to (and rebuttal of) Rule arguably has exercised a much stronger and longer lasting impact on subsequent scholarship than the latter’s original work itself, whilst Southern’s cautious suggestions concerning Eadmer’s scribal identity have no doubt laid the groundwork for several studies published during the last quarter of the twentieth century, particularly by Martin Brett and Neil Ker.19 With regard to Eadmer’s early scribal work, Teresa Webber has provided an authoritative synchronic contextualisation of his activity in the monastic scriptorium of Christ Church, Canterbury during the later eleventh and early twelfth centuries, whilst Michael Gullick’s eminent discussion of Eadmer’s scribal career from a diachronic perspective has revealed significant parallels with the activities of the latter’s fellow twelfth-century monk-historian, Symeon of Durham.20 Whilst these studies have no doubt generated important knowledge about Eadmer’s scribal identity, our understanding of the HNov’s history of composition, codification and dissemination today still rests firmly, and indeed fundamentally, on the foundations laid by Southern.21
MS CCCC 371
Naturally, any reassessment of the manuscripts of Eadmer’s HNov must start with a brief recapitulation of Southern’s hypotheses as regards MS CCCC 371. For our present purposes, these can be summarized as follows: according to Southern, Eadmer composed the HNov in two separate stages, both of which formed an independent and self-contained textual unit.22 The first of these versions Southern believed to have been written in 1109–15, its narrative ending with Anselm’s death on 21 April 1109—a terminating point that appears to be confirmed by the more or less contemporary report of William of Malmesbury referred to earlier in this article. At this stage, the HNov supposedly only comprised Books I–IV, with Books V and VI being appended subsequently following a five-year hiatus in Eadmer’s authorial activity that lasted until at least 1119. To Southern’s mind, Books V and VI thus formed an independent textual unit, secondary to Books I–IV and shaped by an altogether different purpose. He argued that adding these two books had never been part of the work’s original scope and design as envisaged by Eadmer in 1097–1100, when he was travelling around Europe in Anselm’s company and decided, perhaps whilst staying at the abbey of Cluny in Burgundy, to shape his ‘pile of sketches’ into something a little more concrete.23
Southern argued confidently that Books V and VI were conceptualized as something of an ‘afterthought’, which Eadmer added to the existing text once he had returned to his native England after another extended sojourn on the Continent (1116–19) together with Anselm’s successor, Archbishop Ralph d’Escures (1114-22).24 According to Southern, it was not until 1119–25 that Eadmer eventually picked up his pen once again to append Books V and VI to the HNov, at the same time as copying some of his more recent works into MS CCCC 371 along with records of his personal correspondence. These works included not only Eadmer’s account of the miracles of Anselm, but also his treatises on the relics of St Ouen and the conception of the Virgin Mary, his brief account of the life and deeds of Abbot Peter of Canterbury, as well as a dossier of letters that he had exchanged with the Benedictine monks at Worcester and Glastonbury—all in all, a significant scribal effort, for which Eadmer had to insert seven new quires into what was already quite a bulky codex.25 What is of particular interest here is Southern’s argument concerning the contents of MS CCCC 371 before these texts were added, that is, the manuscript’s state of composition when Eadmer and Ralph left England for the Continent in September 1116.26
Southern demonstrated compellingly that at the point of the two men’s departure, Eadmer’s oeuvre collected in his autograph MS CCCC 371 included not just a selection of his shorter works alongside the Vita Anselmi, but also—unlike in its present form—what Southern considered to be the ‘first version’ of the HNov. Once filling the space of ten full quires no longer present in MS CCCC 371 today, the HNov originally occupied the book’s middle section, where it was sandwiched between the treatise ‘On the Happiness of Everlasting Life’ (pp. 261-78) and the Vita Anselmi (pp.299-377).27 The explanation for the removal of these ten quires from MS CCCC 371 put forward by Southern built on his previous argument that Books V and VI represented Eadmer’s ‘second thoughts’, rather than a continuation that had been planned from the outset, and for which Eadmer would have had to reserve blank spaces in his autograph manuscript. Southern argued further that Eadmer removed the ten quires originally accommodating the HNov from MS CCCC 371 precisely so that he could add to it more easily, given that inserting two additional book(let)s into the existing codex supposedly ‘would have made the manuscript very unwieldy’.28 To Southern’s mind, at least, there was ‘no reasonable doubt that these quires [removed from MS CCCC 371 in 1119] contained the Historia Novorum in four books, down to the death of Anselm in 1109. It would, in this state, have just fitted into ten regular quires’.29
As pointed out above, Southern’s reconstruction of the HNov’s history of composition and codification has been adopted more or less wholesale in modern scholarship. Whilst some of his arguments were certainly correct, there are others which can no longer be accepted as accurate in the face of the manuscript evidence. To start with the positive aspects, there is no reason to doubt Southern’s observation that the HNov previously included in MS CCCC 371 occupied the best part of ten quires. Most of the quire numbers still visible at the foot of each quire’s final sheet (verso) are original, having been written by the main scribe, and in the same ink as the main text, which strongly suggests that they were inserted during the original writing campaign. There are certain exceptions, however, namely the quire numbers on pp. 14 (‘·I·’), 22 (‘·II·’), 38 (‘·III·’; see below), 298 (‘XVII·’; see below), 394 (‘XXII·’), 410 (‘XXIII·’), 430 (‘XXIIII·’), 446 (‘XXV·’) and 462 (‘XXVI·’), which were written by a different hand (or perhaps two hands) slightly later than that of the main scribe, but still pointing to the early to mid-twelfth century. In its original state, which Southern dated prior to Eadmer’s departure for Rome in the late summer to early autumn of 1116,30 the manuscript therefore contained a total of thirty- one quires numbered I’ through to ‘XXXI·’ (now pp. 23-378).
The remaining seven quires that today form the beginning (pp. 1-22) and end (pp. 379-462) of MS CCCC 371 are later additions, which also explains why the quire number on p. 38, originally the opening quire of the manuscript, was renumbered from ‘I·’ to ‘·III·’.31 As Southern observed correctly, pp. 298 and 314 now show the same quire number (‘XVII·’).32 The explanation for this duplication is twofold: first, when the ten quires of the HNov originally numbered ‘XVI·’ through to ‘XXV·’ were removed from MS CCCC 371, the next five quires (‘XXVII·’ to ‘XXXI·’; pp. 299-378) were adjusted by careful erasure so as to read ‘[X]XVIP to ‘[X]XXI·’—the five (nota bene: not four) quires that now form pp. 379-462 had not yet been added at this point;33 second, a new quire (pp. 283-298) was inserted into the very space from which the HNov had previously been removed. Given that this quire now followed on directly from the end of quire ‘XV·’ (p.282), it was marked with the number ‘XVII·’, but this time, unlike after the removal of the HNov, the subsequent quire numbers were not adjusted accordingly, which is why the quire number ‘XVII·’ now features twice in MS CCCC 371.
This reconstruction of events is of significance when it comes to establishing the precise contents of the removed quires, and I return to this below. The point at which Southern’s arguments begin to fall short of the manuscript evidence is his confident claim that the ten quires removed from MS CCCC 371 offered the exact space required for Books I-IV of the HNov, and that the text in this state must have corresponded to the work’s first definitive version written in 1109-14. There are several issues arising from this interpretation, each of which becomes evident once we turn to the HNov’s other twelfth-century witnesses, namely MS CCCC 452, MS BL Titus A. ix and the fragment(s) in MS CCCC 341.
MS CCCC 452 AND MS BL TITUS A. IX
As mentioned at the beginning of this article, MS CCCC 452 is both the earliest and most complete version of the HNov to survive from the twelfth century. Written predominantly in Eadmer’s own hand, it contains the entirety of Books I-VI: Praefatio (pp.1–3); Book I (pp.3–77); Book II (pp.77–134); Book III (pp. 134–184), Book IV (pp. 184–259); Book V (pp. 259–341); Book VI (pp. 341–354). It was this manuscript that provided the base text for Rule’s critical edition in 1884, whereas MS BL Titus A. ix—a later twelfth-century copy made from an unknown exemplar—was used primarily to indicate textual variants.34 MS CCCC 452 is still sometimes referred to as Eadmer’s ‘working copy’, even though Brett and others have shown beyond reasonable doubt that the manuscript is in fact a ‘fair copy’ produced after the HNov had been brought to its ultimate point of completion in the early to mid-1120s.35 Before this fair copy can be connected meaningfully to the copy of the HNov that was removed from MS CCCC 371, we must first establish and compare the manuscripts’ physical attributes.
The pages of MS CCCC 452 measure roughly 175 × 118 mm, making them significantly smaller than those of MS CCCC 371 (approx. 226 × 155 mm). To some degree, this difference in size is owing to the fact that MS CCCC 452 was heavily cropped when receiving its current binding, though a comparison of the area reserved for the text itself (approx. 140 × 80 mm) with the equivalent area in MS CCCC 371 (approx. 160 × 100 mm) reveals that MS CCCC 452 was always the smaller of the two manuscripts. The average amount of text on a page of MS CCCC 452 (27 lines per page; avg. 43 letters per line) is considerably smaller than on a page of MS CCCC 371 (33 lines per page; avg. 51 letters per line),36 meaning that a page of MS CCCC 371 holds about one and a half times as much text as a page of MS CCCC 452– or, to put it another way, copying one page of MS CCCC 371 requires about one and a half pages of MS CCCC 452.37 A similar ratio applies with regard to the second twelfth-century copy, MS BL Titus A. ix. Like MS CCCC 452, this manuscript, too, was cropped when it received its modern binding, resulting in measurements of approx. 208 × 145 mm. Fortunately, the writing area remains intact, measuring approx. 156 × 118 mm (34 lines per page; avg. 51 letters per line). This makes the average amount of text copied on a page of MS BL Titus A. ix ever so slightly longer than that on a page of MS CCCC 371.38
Of course, numerical equations such as the ones presented above must always be taken with caution, not least because they operate with averages and thus cannot account for specific variations in a manuscript’s mise-en- page, for example, when inserting initials or other forms of decoration, as well as when compensating for physical imperfections and/or damage such as holes in the parchment, etc. Despite these limitations, the numbers established above are useful in so far as they allow us to revisit Southern’s claim that Books I–IV of the HNov as copied in MS CCCC 452 (barring minor alterations) would have fitted neatly (if not perfectly) into the ten quires removed from MS CCCC 371 before Eadmer appended Books V and VI.39
In MS CCCC 452, Books I–IV (plus the prologue) take up a total of 259 pages (pp. 1–259). Even if we accept Southern’s generalization that the work’s prologue (pp.1–3) was ‘written—like all prefaces—after he [Eadmer] had finished his work’,40—either in 1114 or, less likely, as late as the 1120s–this still leaves 256 pages (or sixteen regular quires).41 Similarly, the HNov’s first four books minus the prologue occupy just under 194 pages (twelve regular quires) in MS BL Titus A. ix (fols. 3v-100v). Transposed on to MS CCCC 371, the same amount of text would have filled the best part of 176 pages (eleven regular quires) (16 × 11 = 176 pages), which is sixteen more pages (one more quire) than we know were in fact removed from the manuscript in or around 1119 (see above). Contrary to Southern’s calculation that the text of Books I-IV originally contained in MS CCCC 371 ‘just fitted into ten regular quires’,42 the evidence provided by MS CCCC 452 and MS BL Titus A. ix suggests that this cannot have been the case if we acknowledge that the length of Books I–IV in these two manuscripts is comparable with—or, when allowing for minor alterations such as those suspected by Southern,43 perhaps slightly longer than—that of Books I–IV originally copied in MS CCCC 371 (or parts thereof). The question must be, therefore, just how much of the HNov’s text as we recognize it today was contained in the ten quires removed from MS CCCC 371?
There are two ways to try to answer this question, the first of which takes us back to the research undertaken by Rule during the late nineteenth century but which was eclipsed subsequently by Southern’s damning criticism. In the preface of his 1884 edition, Rule argued that the text of the HNov as it survives today in MS CCCC 452 and MS BL Titus A. ix does not resemble Eadmer’s original composition of 1109–14, but instead is the result of several consecutive redactions or ‘elaborations’ (to borrow Rule’s terminology) introduced by Eadmer himself during the final years of his life—a hypothesis which he developed further in his follow-up study published in 1886.44 Rule identified a total of thirty-seven textual ‘amplifications’ in Books I–IV, which he classified into eight distinct groups (‘A–H’).45 Four of these (‘A, B, C, F’) he recognized as major ‘recensions’, the earliest of which (‘A’) he dated loosely to the period around the first completed drafts of William of Malmesbury’s Gesta regum Anglorum and Gesta pontificum Anglorum (c. 1125/26).46
Prior to these substantial recensions, Books I–IV in their ‘original state’, dated by Rule to c. 1112, would only have occupied a total of eight regular quires. According to Rule’s interpretation of the HNov’s composition as a gradual process that stretched over the best part of two decades, it was ‘recension A’ that first expanded the text from eight to nine quires around 1125, with ‘recension B’ and ‘recension C’ each adding one more quire during the second half of the 1120s.47 Rule further contended that before ‘recension B’—that is to say, pre-1125—what today is known as the HNov’s Books III and IV had been combined into a single book in Eadmer’s now- lost working copy, and that a preliminary version of what eventually would become Book V (but shorter than its equivalent in MS CCCC 452 and MS BL Titus A. ix) had been added before this split occurred.48 This is an important hypothesis, and I will return to it below.
Unfortunately, it is with the fairly long-winded and laboured reconstruction of Eadmer’s lost working copy that Rule’s arguments quickly abandon their otherwise firm grounding in the manuscript evidence, earning him much criticism by subsequent generations of scholars.49 Southern dismissed the theories of his predecessor out of hand, arguing that they were ‘vitiated by many baseless conjectures’,50 and, as David Rollason has observed, it was Southern’s critique that consigned most of Rule’s scholarship on the HNov to oblivion, including some of its more compelling arguments.51
MS CCCC 341
Moving on from Rule’s ‘conjectures’ and their categorical rebuttal by Southern, I will now offer fresh evidence to establish which parts of the HNov were contained originally in MS CCCC 371, and for this I turn to the final manuscript witness, MS CCCC 341. The main part of the manuscript contains a sixteenth-century copy of the HNov made on the basis of MS CCCC 452, but fixed on to its modern flyleaf are two single-sheet fragments from the twelfth century, one of which contains a passage from the HNov’s Book V, the other a chapter list of the first book of Eadmer’s Vita Anselmi (Figs 1 and 2). Rule was the first to suspect that the twelfth-century codex from which these fragments were removed might have been none other than MS CCCC 371, a suspicion which has been corroborated by Brett.52Indeed, renewed scrutiny of the fragments’ physical and scribal attributes leaves no doubt that they both were once part of MS CCCC 371: both fragments are ruled with dry point, and their layout is a perfect match for that of MS CCCC 371 (33 lines per page; avg. 51 letters per line). The space allocated to the writing area is identical with that in MS CCCC 371 (approx. 160 × 100 mm), and the text itself is written not just by the same hand, but also using what to the naked eye looks like identical ink, which suggests that all three documents were produced during a single (if drawn-out) writing campaign.53
Fragment of the HNov in MS CCCC 341 (recto). Reproduced by kind permission of the Parker Library, Corpus Christi College, Cambridge.
Fragment of the HNov in MS CCCC 341 (verso). Reproduced by kind permission of the Parker Library, Corpus Christi College, Cambridge.
Fragment of the Capitula Vitae Anselmi in MS CCCC 341 (recto). Reproduced by kind permission of the Parker Library, Corpus Christi College, Cambridge.
Fragment of the Capitula Vitae Anselmi in MS CCCC 341 (verso). Reproduced by kind permission of the Parker Library, Corpus Christi College, Cambridge.
The full ramifications of the close physical and codicological relationship between the two fragments in MS CCCC 341 and MS CCCC 371 were not appreciated by Brett when he provided textual and codicological evidence that the fragment of the HNov had probably originally been part of the ten quires removed from MS CCCC 371—and, consequently, that the text in those same quires must have extended beyond the end of the HNov’s Book IV (Anselm’s death in 1109), given that the fragment in fact contains a passage from Book V.54 Indeed, the events recorded on the HNov’s fragment in MS CCCC 341 all occurred in the autumn of 1115, and they include the meeting between King Henry I and the bishops of the realm in the royal palace at Westminster (16 September), the election of the new bishop of St Davids in Wales (18 September) and, importantly, the dispute that ensued over the latter’s consecration the following day (19 September).55 The codicological evidence generated by studying both fragments together, rather than treating them as two separate textual witnesses, allows us to go further than before to demonstrate that the HNov originally contained in MS CCCC 371 must have extended significantly beyond the year 1109, and that both fragments were removed together, and at the same time, from the middle section of MS CCCC 371.
The key here is the second fragment in MS CCCC 341, which lists the chapters of the first book of Eadmer’s Vita Anselmi. Southern believed these capitula to be a rejected draft that Eadmer discarded as obsolete when putting the finishing touches to his autograph copy of Anselm’s Vita in MS CCCC 371 (pp. 299–377).56 Whilst it is true that the wording of some of the chapter headings of the Vita’s opening book now found in MS CCCC 371 (pp. 293–94) shows certain variants compared with the headings written on the fragment in MS CCCC 341, the explanation for the latter’s removal is almost certainly a different one than that suggested by Southern. According to Southern, the full and final list of chapters of the Vita Anselmi surviving in MS CCCC 371 (pp. 293–98) belongs to a series of additions made in 1123–24, which also included a new and expanded ending for the treatise De beatitudine perennis vitae (pp. 278–81) and the Vita Bregowini (pp.281–92).57 There are several problems with this attribution, but these will be discussed in detail elsewhere.58 For now, the task is to determine more precisely why the capitula in MS CCCC 341 were removed from MS CCCC 371 and replaced with those still extant today. If it was not because one was an obsolete draft and the other the revised and final product, then what could be the reason?
The fragments in MS CCCC 341 both show signs of physical damage in the form of a c-shaped hole located about one third down the side of the page that once would have faced the manuscript’s spine. What caused this is hard to determine with certainty, but possible explanations include a glued-on tail band that damaged the pages upon removal. Meanwhile, the same kind of damage also occurs on the last eight pages of MS CCCC 452 (pp. 347–54). Upon closer examination, the fragments in MS CCCC 341 do not just share their ruling pattern, line spacing and page layout with MS CCCC 371 (see above), but they are also of precisely the same physical dimensions as the pages in MS CCCC 452 (approx. 175 × 118 mm). This means that after these fragments had been removed from MS CCCC 371 during the first quarter of the twelfth century, they were bound together with the contents of MS CCCC 452 at some point before the latter had its pages cropped and its binding replaced.
Even though the fragments in MS CCCC 341 today form two separate sheets, they were previously part of a single quire (ternion or quaternion) enclosed between the final page and the back board of MS CCCC 452, now followed by a set of fly- or endleaves made from a double-sheet (pp.402–05).59 This is confirmed beyond doubt by a pair of circular holes/stains on the outer side of the same fly leaves (facing the fore edge) that appear to have been caused by the corroded fastenings of a metal clasp (no longer extant), whose distance (90 mm) and position (14 mm from the edge) are a precise match with an identical pair of holes/stains on the fragments in MS CCCC 341 (Fig. 3). Based on the remarkably regular growth-pattern of the physical damage exhibited in the final quire of MS CCCC 452 (pp. 347–54)—the c-shaped holes grow in size by exactly fifteen per cent with each successive sheet—we can be absolutely certain that the fragment of the HNov was attached directly on to the end of the complete copy of the HNov (Books I–VI) that already existed in the same manuscript, given that the damaged area on the fragment is precisely fifteen per cent larger than that on the last page of Book VI (MS CCCC 452, p. 354) (Figs 4 and 5). By the same principle, the two fragments cannot have been directly adjacent to one another, but they must have been separated by at least two, possibly even four interim sheets (one or two bifolia), thus resulting in a regular quire of either six or eight sheets (twelve or sixteen pages) (Fig. 6).
Holes/stains (from a clasp?) in MS CCCC 341 and MS CCCC 452.
MS CCCC 452, p. 354. Reproduced by kind permission of the Parker Library, Corpus Christi College, Cambridge.
Damage pattern in MS CCCC 452 and MS CCCC 341.
Reconstruction of the original quire XXVI in MS 371.
It seems unlikely that the gathering (re-)used to function as the endleaves of MS CCCC 452 was formed by combining two previously unrelated bifolia taken from different quires of the same source manuscript; rather, we should probably assume that these bifolia were both extracted from a single pre-existing quire which was re-deployed specifically for this purpose. If, therefore, the two fragments in MS CCCC 341 were originally part of the same full quire that contained the final section(s) of the HNov in a version/ recension yet to be identified followed by the chapter list of the Vita Anselmi, and if this quire was amongst the ten quires that we know were removed from MS CCCC 371, then the capitula are not so much a draft rejected on purpose, but rather what we might call ‘collateral damage’. When Eadmer or someone else acting on his behalf decided to remove the HNov from MS CCCC 371, whether to continue the narrative or for some other reason, they could not excise individual pages. Instead, they had to remove entire quires, some of which contained parts of other texts copied directly before and/or after the HNov.
This was less of a problem with regard to the beginning of the HNov, which began neatly on the first page of a new quire (the original quire XVII), and which was preceded by a much shorter text (Eadmer’s De beatitudine perennis vitae) that terminated one and a half pages before the end of the quire (p.281, with the quire number ‘XVP on p.282).60 It was much more problematic, by contrast, with regard to the section that contained the end of the HNov and the beginning of the Vita Anselmi by which it was/is succeeded in MS CCCC 371. Whilst the text of the Vita proper began neatly on the first page of a new quire (p. 299), its capitula had been written on the last six pages of the previous quire (the original quire XXVI). Removing the HNov meant removing the capitula along with it—not by design, but out of necessity. To replace the capitula after their accidental removal and reinsert them in the correct place (before the Vita), a whole new quire had to be inserted, thus leading to the duplication of the quire number ‘XVII.’ in MS CCCC 371 discussed earlier in this article (Fig. 7).
Transmission of the HNov in MSS CCCC 371, 452 and 341.
We can now fill the final gaps and reconstruct the remaining contents of the tenth and final quire (XXVI) removed from Eadmer’s autograph manuscript MS CCCC 371, thereby establishing a more precise chronological endpoint of the HNov originally contained in this manuscript. As we saw above, the quire’s first sheet (*1r–v) was the fragment of the HNov now kept in MS CCCC 341; on its last three sheets (*6r–*8v) were the chapter headings (capitula) of the Vita Anselmi, beginning with the second fragment in MS CCCC 341 (*6r–v). Taken together, this only leaves a limited number of interim sheets unaccounted for: four (*2r–5v) if the quire was a full quaternion, two (*2r–3v) if it was a ternion.61 It seems reasonable to suggest that these interim sheets would have brought the HNov’s narrative to some point of conclusion, if only a provisional one.
This conclusion cannot possibly have been what is known today as the end of Book V, namely Eadmer’s resignation from the episcopacy of St Andrews in Scotland and the death of William Aetheling, King Henry I’s son, in November 1120.62 Based on the average amount of text on a page of MS CCCC 371/341 vis-à-vis MS CCCC 452 calculated earlier in this article, the narrative between the end of the fragment in MS CCCC 341 (September 1115) and the end of Book V in its final form (November 1120) could not possibly have fitted into the remaining space offered by quire XXVI.63 It seems almost certain, therefore, that the HNov originally contained in MS CCCC 371 was an intermediate (and hitherto unrecognised) redaction of the text that Eadmer must have composed in the year or so between September 1115 and his and Ralph’s departure from England in September 1116. As I will show in the remainder of this article, it is possible to narrow down this moment of composition even further.
A new ending?
A cautious calculation based on the average number of letters per page in MS CCCC 371/341 (c. 1,600–1,700, including some allowance for abbreviations, ligatures, contractions, etc.) and projected on to the complete text of Book V copied in MS CCCC 452 (pp. 260–341) and MS BL Titus A. ix (fols. 100v–126r) suggests that the HNov originally contained in MS CCCC 371 ended most likely with Eadmer’s report on the Council of Salisbury summoned by King Henry I on 20 March 1116.64 Whilst this is not the place to offer a detailed analysis of the Council of Salisbury and the way in which it was reported in Eadmer’s HNov, there is some additional evidence to suggest that Eadmer had indeed intended this particular event as an interim point of conclusion (and indeed closure) for his narrative, and that it was recognised and understood as such by his twelfth-century readers.
Eadmer’s report on the Council of Salisbury only occupies a total of fifteen lines in MS CCCC 452 (p.284) and MS BL Titus A. ix (fols. 109v–110r). Rule’s nineteenth-century edition reproduces the text as a single continuous paragraph,65 but the way in which the same passage is formatted in the two twelfth-century manuscripts might indicate that Eadmer and his readers viewed it as two separate episodes. The first of these episodes concludes with the solemn promise by Archbishop Ralph and the other bishops and abbots of the realm to do homage to Prince William Aetheling and accept him as king in the case of Henry’s death.66 Eadmer’s report on these matters is marked as complete with the short phrase ‘De his ita’, which he employs regularly throughout the HNov to signal changes of subject matter.67 Chronologically speaking, it is entirely possible—and indeed likely—that Eadmer composed the text of Book V up to this point as early as April 1116, months before he and Ralph left England together. What follows next, by contrast, is a second and altogether separate episode, an ex post facto report on the drawn-out controversy concerning England’s episcopal primacy (primas totius regni Anglorum) between Ralph and his arch-rival from York, Thurstan.68
Thurstan had been elected as archbishop in 1114 following the death of his predecessor, Thomas II (1109–1114), but he was not consecrated until October 1119 as a result of rather elaborate political manoeuvres–and diplomatic forgeries–that had required both rivals to visit King Henry I in Normandy and Pope Paschal II in Rome.69 Eadmer had accompanied Ralph on these travels, particularly his three-year stint on the Continent in 1116–19 during which Ralph suffered a serious stroke that prevented him from meeting the Pope in person.70 The HNov offers a detailed, if not unbiased, account of the conflict between the ecclesiastical primates of Canterbury and York as it unfolded and escalated during the years 1114–19, and there can be no doubt that Eadmer composed this report with the benefit of hindsight after he had returned to England in 1119. The episode recounting the slow and painful development of this controversy takes up the entire narrative between Eadmer’s report on the Council of Salisbury and what is now recognised as the end of the HNov’s Book V, and it is ushered in with a sentence that draws the reader’s attention to the longevity of the quarrel (querela) .71 Indeed, the episode’s opening sentence (‘Defuncto siquidem Thoma …’) dives in medias res by recounting the disputes that surrounded Thurstan’s election in 1114 and, in particular, the death of his predecessor, Thomas—a cataclysmic event that had given rise to the entire controversy in the first place, and which Eadmer says he has treated at length in the HNov’s previous book.72 The division between this self-contained episode and Eadmer’s previous report on the Council of Salisbury is an important narrative caesura from both a thematic and chronological perspective.
There is additional evidence of this caesura in at least one of the two twelfth-century manuscripts, though certain caveats apply. Beginning with Eadmer’s autograph MS CCCC 452, there is little (if anything) in the manuscript’s mise-en-page to set apart the two episodes, except perhaps for the observation that on p.284 the monochrome initial ‘D’ of Eadmer’s ‘signal phrase’ (‘De his ita’) has been moved to the left-hand tramlines (Fig. 8)—a scribal convention that can be observed frequently throughout the manuscript, and which must be considered standard practice amongst eleventh- and twelfth-century scribes when starting a new sentence on a separate line. Turning to MS BL Titus A. ix, however, we can see that on fol. iior the initial ‘D’ of the second episode’s opening sentence (‘Defuncto …’) has been stroked with red ink, thus visually separating the two episodes from each other (Fig. 9). The significance of this subtle yet effective visual marker is increased further by its uniqueness, given that MS BL Titus A. ix contains no other examples of stroked initials. The sole occurrence on fol. 110r is complemented by the words ‘De Thur[stano]’ (likewise with a stroked initial ‘D’) written in the right-hand margin of the page and penned by a hand contemporary—but most probably not identical—with that of the main scribe, which again remains unparalleled throughout the remainder of the manuscript.
MS CCCC 452, p. 284. Reproduced by kind permission of the Parker Library, Corpus Christi College, Cambridge.
MS BL Titus A. ix, fol. 110r. © The British Library Board
MS BL Titus A. ix is not an autograph, of course, meaning that its idiosyncrasies cannot be considered indicative of Eadmer’s authorial intentions; still, it provides us with important evidence of what we might call ‘reader preference’. The fact that at least one of the HNov’s twelfth-century readers recognized–and indeed marked–the caesura originally introduced by Eadmer is of significance, especially since we do not know from which exemplar MS BL Titus A. ix was copied. It is not impossible, therefore, that similar marks existed in this lost exemplar, too, as well as in some of the other copies of the HNov listed in medieval library catalogues but unaccounted for today. It might be rather telling, in this context, that the fourteenth-century catalogue of books owned by the monks of Christ Church, Canterbury and drawn-up after the death of Prior Henry (de) Eastry (d. 1331) does not just list a ‘Historia novorum in Anglia (l. vi erased)’ and a ‘[H]istoria novorum in Anglia, maior’—the latter of which is MS CCCC 452—but also a ‘[H]istoria Anglorum vetus, libri v’.73 Whilst the most likely candidates for this ‘Historia Anglorum’ are Bede’s Historia Ecclesiastica gentis Anglorum and Henry of Huntingdon’s Historia Anglorum74 we should not rule out the possibility that the ambiguous entry might refer to a lost copy of Eadmer’s HNov—one that, unlike MS CCCC 452 and MS BL Titus A. ix, but similar to MS CCCC 371/341, ended with the events in the spring of 1116 recorded in Book V.
When combining the idiosyncrasies found in MS BL Titus A. ix with the textual analysis provided in this article, as well as with the codicological evidence of the fragments in MS CCCC 341 and the reconstruction of quire XXVI formerly belonging to Eadmer’s autograph MS CCCC 371, a fuller picture emerges. As has become clear, the version of the HNov originally contained in MS CCCC 371 ended neither with the death of Anselm in 1109—the end of Book IV as we know it today—nor with Eadmer’s return from the bishopric of St Andrews and the death of William Aetheling in 1120—now considered the end of Book V. Instead, it ended with Eadmer’s report of the English prelates’ pledge of homage to Prince William at the Council of Salisbury summoned by King Henry I on 20 March 1116, which Eadmer wrote down shortly after the event, but before he left England together with Archbishop Ralph in September that same year. Everything that follows, commencing with Eadmer’s partisan account of the dispute between Ralph and Thurstan of York, constituted a later redaction written after Eadmer’s return to England in 1119. The date of composition of the HNov originally contained in MS CCCC 371 can thus be narrowed down confidently to the six or so months between March and September 1116.
C onclusion
The evidence presented in this article can be interpreted in different ways, and whilst some conclusions are mutually exclusive, they all have significant ramifications for our understanding of the HNov’s history of composition, codification and dissemination during the twelfth century. I conclude my analysis by discussing three possibilities in more detail: the first is that there never was a version of the HNov that comprised only Books I–IV, which would mean that Southern’s widely-accepted reconstruction of a two-stage composition process separated by a five-year hiatus is inaccurate; if Southern was right, however, and if there were two—and no more than two–versions of the HNov during the twelfth century—one with four, the other with six books, a possibility that Brett’s work has rendered altogether unlikely75—then the second possibility is that Eadmer’s autograph originally contained in MS CCCC 371 and surviving in one of the fragments in MS CCCC 341 must have been written post-1119 as part of the six-book version, which would mean that the established terminus ante quem for the composition of MS CCCC 371 is incorrect; (3) finally, if the established date of composition of MS CCCC 371 in its earliest form is correct and there was a four- book version of the HNov, then the work must have been composed, codified and published in a series of intermediate redactions (certainly more than two) between the years 1109 and 1125.
The possibility that, contrary to established scholarly consensus, there never was a four-book version of the HNov seems relatively easy to discard, given that Eadmer himself refers to his early work as ‘written in four books’ on several occasions, and these words are confirmed by his close contemporary, William of Malmesbury.76 Of course, if we follow Southern’s argument that certain sections of historical works such as prologues were usually written last, we might ask whether these references could perhaps reflect a later division of the text which Eadmer implemented after he had finished Books V and VI in the mid-1120s—which once again brings to mind Rule’s suspicion (though discredited by Southern) that ‘the first issue of … Book V was probably (I think, certainly) made prior to the division of the third book of the pristine work into what are now known as Books III and IV’.77 It also seems extremely unlikely that MS CCCC 371 in its earliest form was produced any later than 1119, considering, on the one hand, the respective termini ad quem of its contents established by Southern, and, on the other, the specific ductus of Eadmer’s hand that penned large parts of this manuscript, and which resembles his hand in MS CCCC 452 much more closely than in the various books and documents which Gullick and Webber have identified as belonging to the earlier phases of his scribal career up to 1109.78
The most likely explanation, therefore, is this: like most twelfth-century writers of history, Eadmer did not compose his HNov in two sessions separated by half a decade or so of inactivity, but rather in a series of redactions which he wrote, rewrote and revised over extended periods of time. The ways in which MSS CCCC 371, 452 and 341 were assembled, dismembered and reassembled over several decades demonstrate that Eadmer continuously modified and adjusted his autograph manuscripts as he progressed. Naturally, this working method generated a variety of different redactions, only a few of which have survived today. Whilst future study might reveal just how many different redactions of the text were in circulation during the twelfth century, the reassessment of the HNov’s earliest manuscripts offered in this article has shown that one such redaction was certainly completed and codified in MS CCCC 371 between March and September 1116.
This, then, is the version of the HNov that was removed from MS CCCC 371 around 1119 together with the capitula of the Vita Anselmi—which must be considered ‘collateral damage’, rather than an obsolete draft—meaning that its only surviving traces, to the best of my knowledge, are the two fragments of quire XXVI that were bound into MS CCCC 452 after their removal from MS CCCC 371, and which today constitute the flyleaf of MS CCCC 341 (Fig. 7). If there were a hiatus in the writing of the HNov, this interruption surely must have been relatively short-lived, being limited to the three years Eadmer spent travelling around Europe in the company of Archbishop Ralph. Even then, it seems entirely plausible that Eadmer continued to take notes and produce drafts whilst en route–just as he had done on his journeys with Anselm twenty years earlier—and that he transferred these drafts into his fair autograph MS CCCC 452 not long after his return to England in 1119. Before bidding farewell to his brothers at Christ Church, Canterbury in the summer of 1116, Eadmer had decided to ‘publish’ the most recent version of the HNov that ended with the Council of Salisbury in MS CCCC 371. It seems unlikely that he had intended for this text to be disseminated outside of his own monastic community (we might thus speak of ‘internal publication’), though of course in reality he had little control over how it was read, lent, borrowed or copied during the period of his absence. Only upon his return to Canterbury in 1119 did Eadmer get another chance to interfere with the book he had left behind three years earlier, which he did by removing the HNov and adding the content he had drafted in the interim, thus producing the next redaction of what would become one of the most important history books of the twelfth century.
Footnotes
My research on the manuscripts that form the basis of this article was kindly facilitated by the Parker Library, Corpus Christi College, Cambridge, particularly by its sub-librarian, Anne McLaughlin, for whose support I am very grateful. My thanks also go to Samu Niskanen and Keith Busby for reading an early draft, and to Charlie Rozier and Levi Roach for offering their advice on different aspects of the argument. Last but not least, I am indebted to The Library's anonymous peer- reviewers for their constructive comments and suggestions. All remaining errors are, of course, entirely my own.
1 Cambridge, Corpus Christi College, MS 452, p. 1: [V]ideor mihi videre magnum quid posteris praestitisse, qui suis gesta temporibus, futurorum utilitati studentes, litterarum memoriae tradidere … Hoc igitur considerato penes me, statui ea quae sub oculis vidi vel audivi, brevitati studendo, stili officio commemorare. A digitized version of the manuscript is available at https://parker.stanford.edu/ parker. The text is available in Eadmeri Historia novorum in Anglia et opuscula duo de vita Sancti Anselmi et miraculis eius, ed. by Martin Rule (London: Longman, 1884).
2 The precise dates of Eadmer’s life are unknown, though he is widely accepted to have been born around 1060 and died between 1126 and 1128. See Jay Rubenstein, ‘Eadmer [Edmer] of Canterbury’ in ODNB. See also Richard W. Southern, Saint Anselm and His Biographer: A Study of Monastic Life and Thought, 1059–c.1130 (Cambridge: Cambridge University Press, 1963), pp.229–40; Eadmer of Canterbury: Lives and Miracles of Saints Oda, Dunstan, and , ed./tr. by Bernard J. Muir and Andrew J. Turner (Oxford: Clarendon Press, 2006), pp.xiii–xxxv; Bernd Goebel, Im Umkreis von Anselm: Biographisch-bibliographische Porträts von Autoren aus Le Bec und Canterbury (Wurzburg: Echter, 2017), pp. 31–67.
3 Eadmer’s vision of history has been explored by Jay Rubenstein, ‘Liturgy against History: The Competing Visions of Lanfranc and Eadmer of Canterbury’, Speculum, 74 (1999), 279–309. See also Charles C. Rozier, ‘Between History and Hagiography: Eadmer of Canterbury’s Vision of the Historia novorum in Anglia’, Journal of Medieval History, 45 (2018), 1–19. On the relationship between history and hagiography (and liturgy) in the age of Eadmer, see Susan Boynton, ‘Writing History with Liturgy’, in Representing History, 900–1300: Art, Music, History, ed. by Robert A. Maxwell (University Park, PA: Penn Press, 2010), pp.187–200; Steven Vanderputten, ‘Typology of Medieval Historiography Reconsidered: A Social Re-interpretation of Monastic Annals, Chronicles and Gesta’, Historical Social Research, 26 (2001), 141–78; Tom Licence, ‘History and Hagiography in the Late Eleventh Century: The Life and Work of Herman the Archdeacon, Monk of Bury St Edmunds’, English Historical Review, 124 (2009), 516–44. See also the various contributions to Medieval Cantors and Their Craft: Music, Liturgy and the Shaping of History, 800–1500, ed. by Katie A.-M. Bugys, Andrew B. Kraebel and Margot E. Fassler (Woodbridge: Boydell, 2017). On Anselm’s relationship with Eadmer, see Sally N. Vaughn, ‘Anselm of Le Bec and Canterbury: Teacher by Word and Example, Following the Footprints of His Ancestors’, in A Companion to the Abbey of Le Bec in the Central Middle Ages (11th–13th Centuries), ed. by Benjamin Pohl and Laura L. Gathagan (Leiden: Brill, 2017), pp. 57–93.
4Eadmeri monachi Cantuariensis vita sancti Anselmi archiepiscopi Cantuariensis = The Life of St Anselm, Archbishop of Canterbury by Eadmer, ed. and tr. by Richard W. Southern (Oxford: Clarendon Press, 1972); Vite di Anselmo d’Aosta / Eadmero e Giovanni di Salisbury, tr. by Inos Biffi and others (Milan: Jaca Book, 2009). The principal study on the Vita Anselmi remains Southern, Saint Anselm and His Biographer. See also Michael Staunton, ‘Eadmer’s Vita Anselmi: A Reinterpretation’, Journal of Medieval History, 23 (1997), 1–13; Simon McGurk, ‘Anselm through the Eyes of Eadmer’, in A Man Born out of Due Time: New Perspectives on St Anselm of Canterbury, ed. by Dunstan Robidoux (New York, NY: Lantern, 2013), pp.1–32. Eadmer’s other works include the Vita Sancti Wilfridi auctore Edmero = The Life of Saint Wilfrid by Edmer, ed. and tr. by Bernard J. Muir and Andrew J. Turner (Exeter: University of Exeter Press, 1998), as well as a number of shorter works published in Lives and Miracles, ed. and tr. by Muir and Turner. For a full list of Eadmer’s oeuvre, see Richard Sharpe, A Handlist of the Latin Writers of Great Britain and Ireland before 1540 (Turnhout: Brepols, 1997), pp. 59–61. Also cf. Goebel, Im Umkreis von Anselm, pp.45–52.
5 Richard Southern, St Anselm: A Portrait in a Landscape (Cambridge: Cambridge University Press, 1990), P.405.
6 William of Malmesbury, Gesta regum Anglorum = The History of the English Kings, ed. and tr. by Rodney M. Thomson and Michael Winterbottom, 2 vols (Oxford: Clarendon Press, 1998–99), 1, 14–15: ‘Nec vero nostrum effugit conscientiam domni Edmeri sobria sermonis festivitate elucubratum opus, in quo … licentius evagatus usque ad obitum Anselmi archiepiscopi diffusam et necessariam historiam studiosis exhibuit’.
7 ‘Ita pretermissis a tempore Bedae ducentis et viginti annis, quos iste nulla memoria dignatus est, absque litterarum patrocinio claudicat cursus temporum in medio’; ibid. Cf. Benjamin Pohl, Dudo of Suint-Quentin’s Historia Normannorum: Tradition, Innovation and Memory (York: York Medieval Press, 2015), pp.16–17; Michael Staunton, The Historians of Angevin England (Oxford: Oxford University Press, 2017), p.26.
8Historia novorum in Anglia, ed. Rule. Rule faithfully reproduced the Latin text of the HNov divided into six books from the two principal manuscript witnesses (see below), but only four of these books have since been published in translation. Eadmer’s History of Recent Events in England = Historia novorum in Anglia, tr. by Geoffrey Bosanquet (London: Cresset, 1964) only includes Books I–IV, and the same is true of the subsequent French and Italian translations: Histoire des temps nouveaux en angleterre (livres I–IV) / Eadmer, moine de Cantorbéry, tr. by Henri-Marie Rochais (Paris: Éditions du Cerf, 1994); Historia novorum in Anglia = Storia dei tempi nuovi in Inghilterra, Libri I–IV, tr. by Antonio Tombolini (Milan: Jaca Book, 2009).
9 The one exception, in this regard, is the insightful note by Martin Brett, ‘A Note on the Historia Novorum of Eadmer’, Scriptorium, 33 (1979), 56–58. Studies that contextualize Eadmer’s manuscripts within the scribal culture of Christ Church, Canterbury include Teresa Webber, ‘Script and Manuscript Production at Christ Church, Canterbury, after the Norman Conquest’, in Canterbury and the Norman Conquest: Churches, Saints, and Scholars, 1066–1109, ed. by Richard Eales and Richard Sharpe (London: Hambledon, 1995), pp.145–58; Michael Gullick, ‘The Scribal Work of Eadmer of Canterbury to 1109’, Archaeologia Cantiana, 118 (1998), 173–89. See also T. A. M. Bishop, ‘Notes on Cambridge Manuscripts: Part I’, Transactions of the Cambridge Bibliographical Society, 1 (1953), 432–41; N. R. Ker, ‘Copying an Exemplar: Two Manuscripts of Jerome on Habakkuk’, in Miscellanea Codicologica F. Masai Dicata MCMLXXIX, ed. by Pierre Cockshaw, Monique-Cécile Garand, and Pierre Jodogne, 2 vols (Ghent: E. Story-Scientia S.P.R.L., 1979), 1, 203–10.
10 For example, Sally N. Vaughn, ‘Eadmer’s Historia Novorum: A Reinterpretation’, Anglo-Norman Studies, 10 (1988), 259–89 (p.287); Southern, Portrait in a Landscape, pp.435–36, 470–73. However, there now is evidence to suggest that William might have had access to a copy of the at Malmesbury, and that he relied on this copy, rather than the one at Canterbury, for the composition of London, Lambeth Palace Library, MS 224–William’s partial autograph, written c. 1119–25, in which he recorded rare details about Anselm’s career at Le Bec based on a combination of Anselm’s letters and Eadmer’s HNov. I owe this information to Samu Niskanen, who kindly shared it with me prior to the publication of his edition, Letters of Anselm, Archbishop of Canterbury, Volume 1 (Oxford: Clarendon Press, forthcoming). The HNov also provided a source for Gervase of Canterbury’s Gesta regum and Actus pontificum Cantuariensis ecclesiae, written around the turn of the thirteenth century, and it seems reasonable to assume that Gervase saw a copy of the work at Christ Church, Canterbury; see Lives and Miracles, ed. and tr. by Muir and Turner, pp. liv–lvii
11 In keeping with the work of Richard Sharpe, I am using the term ‘(to) publish’ in the sense of ‘(to) give out to the public what has been written’; see Richard Sharpe, ‘Anselm as Author: Publishing in the Late Eleventh Century’, Journal of Medieval Latin, 19 (2009), 1–87 (p. 1). For a discussion of medieval publishing practices based primarily (though not exclusively) on vernacular manuscripts, see Leah Tether, Publishing the Grail in Medieval and Renaissance France (Cambridge: D. S. Brewer, 2017), pp.13–26. Also cf. the ongoing ERC-project ‘Medieval Publishing from c.1000 to 1500’ at the University of Helsinki, https://www.helsinki.fi/en/researchgroups/medieval-publishing, whose principal investigator, Samu Niskanen, kindly invited me to present a keynote based on an early version of this article at the symposium ‘Publishing and Paratext in the Medieval and Early Modern Periods’ (Helsinki 11–12 June 2018). In revising the article for publication, I have benefited greatly from the helpful comments and suggestions I received from the audience. My particular thanks go to Jaakko Tahkokallio and Matti Peikola.
12 Richard Sharpe and James Willoughby in their electronic resource ‘Medieval Libraries of Great Britain’ (MLGB3) list two copies of HNov kept at Christ Church, Canterbury during the Middle Ages (BC4.188; BC4T189), http://mlgb3.bodleian.ox.ac.uk/, both of which are recorded in the catalogue of books drawn up during the priorate of Henry (de) Eastry (1284/85–1331); see Montague R. James, The Ancient Libraries of Canterbury and Dover (Cambridge: Cambridge, University Press, 1903), p. 37 (= nos. 188† and 189). One of these (BC4. *189) is MS CCCC 452, whereas the identity of the other (BC4.188), which is assumed lost, has not been established yet.
13 See below for a detailed discussion of these manuscripts and the relevant literature.
14 The presence of hands other than Eadmer’s was noticed by Southern, Saint Anselm and His Biographer, p. 372, who remarked upon a change of hands in MS CCCC 452, p. 298 (occursus et seqq.).
15 The British Library’s online catalogue gives the date as ‘2nd half of the 12th century’. Also cf. the entries in Catalogus librorum manuscriptorum bibliothecae Cottonianae, ed. by Thomas Smith (Oxford: Sheldonian Theatre, 1696), p. 122; A Catalogue of the Manuscripts in the Cottonian Library Deposited in the British Museum, ed. by Joseph Planta (London: Hansard, 1802), p. 512. This is not the place to resolve the question as to whether MS BL Titus A. ix is in fact the manuscript recorded in the possession of Haughmond abbey, Shropshire by John Leland during the first half of the sixteenth century; see Historia novorum in Anglia, ed. Rule, p. xv; Elisabeth M. C. van Houts, ‘Camden, Cotton and the Chronicles of the Norman Conquest of England’, The Electronic British Library Journal (1992), 148-62 (p.158).
16 Both Rule and Southern were aware of this fragment; see Historia novorum in Anglia, ed. Rule, pp. xi–xii; Southern, Saint Anselm and His Biographer, p. 372. The most detailed study to date remains Brett, ‘Note on the Historia Novorum’
17 Martin Rule, ‘Eadmer’s Elaboration of the First Four Books of his Historia Novorum’, Proceedings of the Cambridge Antiquarian Society, 6 (1886), 194–304.
18 Southern, Saint Anselm and His Biographer, pp. 367–74 (= ‘Appendix III: Eadmer’s Personal Manuscript’) (pp. 371–74 for the identification of Eadmer’s hand). Also cf. Southern, A Portrait in a Landscape, p.421.
19 Brett, ‘Note on the Historia Novorum’; Ker, ‘Copying an Exemplar’.
20 Webber, ‘Script and Manuscript Production’, pp. 148–52; Gullick, ‘Scribal Work of Eadmer’, р. 186. Both these studies focus exclusively on Eadmer’s early scribal activity prior to 1109, meaning that they can shed no light on the production of the HNov’s autograph manuscripts. On the similarities between the careers of Eadmer and Symeon, see also David Rollason, ‘Symeon of Durham’s Historia de regibus Anglorum et Dacorum as a Product of Twelfth-Century Historical Workshops’, in The Long Twelfth-Century View of the Anglo-Saxon Past, ed. by Martin Brett and David A. Woodman (Aldershot: Ashgate, 2015), pp.95–112. Symeon’s career as a ‘cantor-historian’ has been studied by Charles C. Rozier, ‘Symeon of Durham as Cantor and Historian at Durham Cathedral Priory с.1090–1129’, in Medieval Cantors and Their Craft, ed. Bugys, Kraebel and Fassler, pp. 190–206. On Symeon’s handwriting, see Michael Gullick, ‘The Scribes of the Durham Cantor’s Book (Durham, Dean and Chapter Library, MS B.IV.24) and the Durham Martyrology Scribe’, in Anglo-Norman Durham, 1093—1193, ed. by David W. Rollason and others (Woodbridge: Boydell, 1994), pp.93–109; ‘The Hand of Symeon of Durham: Further Observations on the Durham Martyrology Scribe’, in Symeon of Durham: Historian of Durham and the North, ed. by David Rollason (Stamford: Shaun Tyas, 1998), pp. 14–31.
21 On Southern’s own partial revision of his previous hypotheses, motivated not least by the work of Sally Vaughn, see Samu Niskanen, ‘Review of Archbishop Anselm, 1093–1109: Bec Missionary, Canterbury Primate, Patriarch of Another World, by Sally N. Vaughn’, Journal of Ecclesiastical History, 65 (2014), 169–72 (p. 170).
22 On what follows, cf. Southern, Saint Anselm and His Biographer, pp. 368–69; Southern, Portrait in a Landscape, pp. 412–14.
23 See ibid. p.412, where it is suggested that Eadmer by that point ‘must have had quite a pile of sketches … on scraps of parchment, on wax tablets, on loose quires almost in book form. He had perhaps by now decided to make two works out of his accumulated notes: one on Anselm’s Life [= Vita Anselmi]; the other on the public events which he had witnessed [= HNov].’ It is not entirely clear from Southern’s discussion whether this ‘pile of sketches’ is supposed to have included both the HNov and the Vita Anselmi or merely the latter.
24 ibid. pp.415–18. In his foreword to Geoffrey Bosanquet’s translation of the HNov, Southern argued that ‘the second thoughts [represented by Books V–VI] are very easily distinguishable. The work originally stopped promptly with the death of Anselm and its immediate consequences . In this form [= Books I–IV] the work was finished soon after Anselm’s death. The additions which were made ten years later had a rather different purpose’; Eadmer’s History of Recent Events, tr. Bosanquet, p. xi. Eadmer’s travels in 1116–19 are discussed later in this article. On Ralph d’Escures, see Jean Truax, Archbishops Ralph d’Escures, William of Corbeil and Theobald of Bec: Heirs of Anselm and Ancestors of Becket (Farnham: Ashgate, 2012), pp.27–74; Mary A. Clark, ‘Ralph d’Escures: Anglo-Norman Abbot and Archbishop’ (unpublished PhD diss., University of California at Santa Barbara, 1975); also Martin Brett’s article in ODNB.
25 Southern distinguished four different groups (‘Groups A–D’) of additions made by Eadmer to MS CCCC 371, and he dated them to c.1120–21 (A), c.1123–24 (B–C) and c.1124–30 (D), respectively; Southern, Saint Anselm and His Biographer, pp. 369–71.
26 Truax, Archbishop Ralph d’Escures, pp.49–74; Goebel, Im Umkreis von Anselm, pp.34–35; Southern, Portrait in a Landscape, p.417.
27 Southern, Saint Anselm and His Biographer, p. 368; Rubenstein, ‘Eadmer of Canterbury’.
28 Southern, Saint Anselm and His Biographer, p. 368.
29 ibid. Also cf. Southern, Portrait in a Landscape, p. 413.
30 Southern, Saint Anselm and His Biographer, p. 368.
31 Whether these were all added at once or at different points in time, as suggested by Southern (ibid. pp. 369–71), is difficult to determine. I am currently preparing a separate study which will scrutinize these quires’ sequence of composition (and that of the texts they contain) in greater detail.
32 ibid. p. 370.
33 Southern mistakenly stated that ‘four new quires were added at the end of the volume’ (ibid.).
34 Prior to Rule’s edition, MS BL Titus A. ix had been used as the principal witness of the HNov ever since the seventeenth century, when it formed the basis of the text’s editio princeps by John Selden; Eadmeri monachi Cantuariensis historiae novorum sive sui saeculi libri VI, ed. by John Selden (London: William Stanesbey, 1623).
35 Brett, ‘Note on the Historia Novorum’, pp. 56–57; also Historia novorum in Anglia, ed. Rule, p. xiv.
36 Note that the first quire in MS CCCC 452 (pp. 1–16) only has 24 lines per page.
37 Based on the exact figures and measurements presented above, one page of text in MS CCCC 371 (avg. 1,680 letters) thus equates to approx. 1.45 pages of text in MS CCCC 452 (avg. 1,160 letters).
38 One page in MS BL Titus A. ix (avg. 1,730 letters) equals approx. 1.05 pages in MS CCCC 371, and approx. 1.49 pages in MS CCCC 452. Note, however, that there is some variation in the number of lines per page throughout the first quire (fols. 3r–10v) of MS BL Titus A. ix, as fols. 3r and 9r have 33 lines each, and fol. i0v only 32 lines.
39 See Southern, Saint Anselm and His Biographer, p. 299, where he discusses the possibility that ‘there are two or three passages which have every appearance of being additions, and it is clear that the last two pages of Book IV were added when Eadmer took up the work again in 1119’. I demonstrate below, however, that these minor alterations as allowed for by Southern could not possibly have brought down the work to the length of the ten quires removed from MS CCCC 371.
40 Southern, Portrait in a Landscape, p. 415. It is not entirely clear whether Southern thought Eadmer wrote the HNov’s prologue upon completion of Books I–IV, or after the addition of Books V and VI a decade or so later. Also cf. Antonia Gransden, ‘Prologues in the Historiography of Twelfth-Century England’, in England in the Twelfth Century: Proceedings of the 1988 Harlaxton Symposium, ed. by Daniel T. Williams (Woodbridge: Boydell, 1990), pp. 55–81.
41 In a manuscript of regular quaternions (four double-sheets = 16 pages), these 256 pages could be bound into exactly 16 quires (16 × 16 = 256 pages). Note, however, that the quire structure of MS CCCC 452 is slightly irregular: whilst the majority of the manuscript is bound in quires of eight sheets (four bifolia), quire nos. 7 (pp.97–114), 9 (pp.135–152) and 21 (pp.337–354) consist of nine sheets, whereas quire nos. 8 (pp. 115–134), 16 (pp. 249–268) and 18 (pp. 285–304) have ten sheets (five bifolia). This variation in quire size throughout MS CCCC 452 opens up the possibility, at least in theory, that some of the ten quires removed from MS CCCC 371 likewise may have comprised more or less than four bifolia, though in the absence of these quires this remains impossible to know.
42 Southern, Saint Anselm and His Biographer, p. 368.
43 ibid. p. 299.
44 Rule, ‘Eadmer’s Elaboration’, passim.
45 See the lists in ibid. pp. 202–3, 235–36.
46 ibid. pp.302–03; Historia novorum in Anglia, ed. Rule, pp.lv–lvi. On the date of William of Malmesbury’s draft of the Gesta regum Anglorum (Wi), see Gesta regum Anglorum, ed. and tr. Thomson and Winterbottom, ii, xxii–xxiii. It has been argued that this draft version was marked up specifically so as to serve as the exemplar for a royal presentation copy intended for the Empress Matilda; see the discussion by Bruce R. O’Brien, ‘Review of Gesta regum Anglorum = The History of the English Kings, ed. and tr. by Rodney M. Thomson and Michael Winterbottom’, Speculum, 76 (2001), 815–18 (p.816).
47Historia novorum in Anglia, ed. Rule, pp.xl–xli, lxxx; Rule, ‘Eadmer’s Elaboration’, p. 302.
48 ibid.
49 Particularly problematic is Rule’s over-confident attempt to calculate the precise amount of text on individual pages of this working copy (often down to the specific number of lines and the number of letters contained therein), for which he establishes an artificial base unit of twenty-four and a half lines (or even multiples thereof); see ibid. pp. 201–04.
50 Southern, Saint Anselm and His Biographer, p. 372.
51 Rollason, ‘Symeon of Durham’s Historia de regibus Anglorum’, p. 109.
52Historia novorum in Anglia, ed. Rule, pp.xi–xii; Brett, ‘A Note on the Historia Novorum’, p. 57.
53 Forensic analysis of the ink might bring further clarity here.
54 Brett, ‘A Note on the Historia Novorum’, p. 58.
55 The passage from Book V contained on the fragment in MS CCCC 341 is Historia novorum in Anglia, ed. Rule, pp.233–35 (… permissum est… consecret …).
56 Southern, Saint Anselm and His Biographer, p. 372.
57 Southern subsumed these additions under ‘Group C’; see ibid. p. 370.
58 For example, the manuscript clearly already contained an ending of De beatitudine perennis vitae of the same length, as is evidenced by a series of erasures. For further discussion, I refer the reader to my forthcoming study (co-authored with Leah Tether) on the publication of Eadmer’s autograph manuscripts in a thematic volume edited by Sally Vaughn and Charlie Rozier under the working title Eadmer of Canterbury: Historian, Hagiographer, and Theologian.
59 These flyleaves might have been followed by a pastedown, but this cannot be known for sure. The group of blank paper leaves that now separates the end of the HNov from the flyleaves in MS CCCC 452 is almost certainly a subsequent addition.
60 As Southern observed, the Vita Bregowini is a later addition written into the blank spaces after De beatitudine perennis vitae and extending into the new quire that had been inserted after the removal of the HNov to replace the capitula; cf. Southern, Saint Anselm and His Biographer, p. 370.
61 It seems highly unlikely that the quire was a quinion (five bifolia), given that no other quinions are found anywhere in MS CCCC 371, whose quire structure—apart from occasional additions and/or removals such as those discussed above—is remakably regular; cf. the metadata provided by the Parker Library on the Web viewer, which gives the collation of MS CCCC 371 as follows: I(8) (1 canc.) II (4) || I(8)-V(8) (+1) VI(8)-XIII(8) (+ slip after 2) XIV(8)-XXIV(8) XXV(10) XXVI(8) XXVII(8).
62Historia novorum in Anglia, ed. Rule, pp. 286–89.
63 This remains true even in the unlikely case that the quire was a quinion (see above).
64 ibid. p. 237. On the historical context, see Truax, Archbishop Ralph d’Escures, pp.41–45.
65Historia novorum in Anglia, ed. Rule, pp.237–38.
66 ibid. p. 237: ‘Radulfus autem archiepiscopus Cantuariensis et alii episcopi atque abbates regni Anglorum fide et sacramento professi sunt se et regnum et regni coronam, si, defuncto patre suo, superviverent, in eum, omissa omni calumnia et occasione, translaturos, eique, cum rex foret, hominia fideli mente facturos.’
67 See, for example, Historia novorum in Anglia, ed. Rule, pp. 18, 181, 278, 288.
68 On this conflict, see Truax, Archbishop Ralph d’Escures, pp.42–47; Clark, ‘Ralph d’Escures’, pp. 134–5 3. Also cf. Denis Bethell, ‘William of Corbeil and the Canterbury York Dispute’, Journal of Ecclesiastical History, 19 (1968), 145–59.
69 On these diplomatic forgeries, see particularly Robert F. Berkhofer III, ‘The Canterbury Forgeries Revisited’, Haskins Society Journal, 18 (2006), 36–50.
70 Truax, Archbishop Ralph d’Escures, pp. 54–55; Southern, A Portrait in a Landscape, pp.416–17.
71Historia novorum in Anglia, ed. Rule, p. 237: ‘Habita quoque est his diebus causa de querela quae inter archiepiscopum Cantuariorum et electrum pontificem Eboracensem per integrum pene annum versata fuerat.’
72 Ibid.: Defuncto siquidem Thoma, cuius circa finem quarti libri supra memoratae Historiae satis habita mentio est, electus erat ad regimen praedicti pontificates quidam de clericis regis vocabulo Thurstanus, convivente Radulfo archiepiscopo, et aliam quam rei exitus probavit de eo habente opinionem.’
73 James, Ancient Libraries, p. 37 (= nos. 187, 188† and 189). Also cf. the entries in MLGB3.
74 Bede’s Ecclesiastical History of the English People, ed. and tr. by Bertram Colgrave and R. A. B. Mynors, rev. edn (Oxford: Clarendon Press, 1992); Henry, Archdeacon of Huntingdon: Historia Anglorum = The History of the English People, ed. and tr. by Diana Greenway (Oxford: Clarendon Press, 1996).
75 Brett, ‘Note on the Historia Novorum’, pp. 57–58.
76Historia novorum in Anglia, ed. Rule, pp.217–18, 237; Gesta regum Anglorum, ed. and tr. Thomson and Winterbottom, 1, 14–15.
77 Rule, ‘Eadmer’s Elaboration’, p. 200. Cf. Southern, Portrait in a Landscape, p. 415.
78 Southern, Saint Anselm and His Biographer, p. 368: ‘[I]t is a reasonable inference that the manuscript [MS CCCC 371] in its original state was in existence by this date [1116] at the latest’; Gullick, ‘Scribal Work of Eadmer’, pp. 174–75; Webber, ‘Script and Manuscript Production’, pp. 149–50.
