Using Decorative Elements to Refine the Relationship Between Two Ælfrician Manuscripts

Written by: Dr Emily Rae (University of Glasgow)

Corresponding email: e.rae.1@research.gla.ac.uk

Abstract

Download as PDF

© 2021 Emily Rae
This work is licensed under the Creative Commons Attribution 4.0 International License


Items from Ælfric’s First Series of Catholic Homilies survive in thirty-four manuscripts and manuscript fragments. Because of their complex histories, and our incomplete modern knowledge of their production and dissemination, we still have only a limited understanding of the exact relationship between all copies of the First Series texts. In the past, scholarship has determined manuscript relationships based primarily on textual collation, rather than by considering physical aspects of the manuscripts themselves. This article demonstrates that attention to extra-linguistic aspects of these manuscripts in relation to decoration and mise-en-page can help to qualify our understanding of these relationships. I here look at two manuscripts containing primarily Ælfrician texts — Oxford, Bodleian Library, Bodley 340 and 342, and Cambridge, Corpus Christi College 162 — and argue that these two manuscripts are not only related, but perhaps very closely so, even sharing a direct exemplar-descendent relationship. I base my argument both on the texts shared between the manuscripts and aspects of the decoration that are uniquely similar between the two.

Introduction

The works of Ælfric comprise the majority of the corpus of surviving Old English prose. In the last few decades, the manuscripts housing these writings have begun to attract increased attention as objects of study in their own right, in part due to the growing scholarly interest in manuscript context; in more recent years, scholars have turned their attention to investigating how this context, aside from its relevance for the study of manuscripts as physical objects, can bolster our understanding of the texts found within. Elaine Treharne, Mary Swan, Orietta Da Rold, and others have all made invaluable contributions to the field of specifically Ælfrician manuscript studies. Much of Treharne’s work has focused on how later readers of Ælfrician manuscripts understood and interacted with the texts and manuscripts;1 alongside online collaborative projects on later Old English manuscripts,2 this body of scholarship has provided much of the foundation for this current work on how contemporary readers and scribes interacted with and copied the texts.

In most previous scholarship on Ælfrician manuscripts, the focus of the study is either the manuscript itself as an objet d’art, or the work is an analysis of the textual content within. It is less common that the two approaches are used in tandem with one another, despite the clear benefits of such a combination. By treating both approaches as two complementary parts of a whole, we are able to create a much fuller reconstruction of the complicated chains of manuscript and textual transmission that would otherwise be lost to us today. In this article, I use a combination of the two methods in order to investigate two Ælfrician manuscripts, as well as drawing attention to heretofore unnoticed similarities between them. This comparison both adds further context to the currently posited relationship between these volumes, as well as suggests a new method of inquiry into all of the manuscripts that contain Ælfrician content.

In his ground-breaking edition of the First Series of Ælfric’s Catholic Homilies, Peter Clemoes based his texts on a single authoritative manuscript from the late-tenth century, British Library, Royal MS 7 C XII, a manuscript widely considered to have been created under Ælfric’s direct supervision.3 He assigned a siglum to every manuscript or manuscript fragment that included any text from Ælfric’s First Series, thirty-four in total.4 His series of stemmata show how versions of the First Series texts may have been distributed and arranged, divided into six separate phases of change and dissemination.5 These stemmata do, on occasion, represent direct connections between extant manuscripts: for example a text in Cambridge, Corpus Christi College 178 can be identified as the direct exemplar for the text in another collection, Oxford, Bodleian Library, Hatton 113+114 and Junius 121.6 Clemoes determined this relationship with textual evidence: revisions that were made in the margins of CCCC 178 are subsequently adopted in Hatton 113+114 and Junius 121, but appear in none of the other Ælfrician manuscripts.7 Much more frequently, however, the textual history of a manuscript is considerably more complicated than this, and straightforward relationships can only very rarely be posited based on evidence of this type. Most of the time, Ælfrician manuscripts drew upon multiple sources during production; it is only very infrequently that we find possible instances of new manuscripts being copied from only a single exemplar. Even in an instance where Clemoes has claimed that the Ælfrician texts in Cambridge, Corpus Christi College 303 were all copied from a single source, this proposed relationship has proven controversial, as Kathryn Lowe claims that in fact closer textual analysis shows that the two parts of the manuscript were copied from separate exemplars.8 Adding an additional level of complication is the fact that many of the codices that almost certainly existed at some point in the past are now simply lost. Clemoes accounts for these in the stemmata by including unlabeled nodes that represent posited manuscripts, and is thus able to argue, for example, that two specific manuscripts likely directly descended from the same now-missing exemplar. As these hypothesized manuscripts necessarily no longer exist, any information about them is at best an educated guess based on extrapolating backwards from their descendants. All of the above issues have resulted in a limited understanding of how all of the Ælfrician manuscripts may connect to one another, as well as how the physical composition of the manuscripts may concern the information contained within them. As mentioned above, individual scholars have recently begun to focus more on this issue, working with a small number of manuscripts at a time to create a more refined image of how these manuscripts may be related to one another.

The Manuscripts

The two volumes I focus on here are Oxford, Bodleian Library, Bodley 340 and 342 (Bodley 340+342) and Cambridge, Corpus Christi College 162, pp. 1–138 and 161–564 (CCCC 162). In his edition of Ælfric’s First Series, Clemoes concluded that these two manuscripts are not especially closely related to one another: while both manuscripts are at least partially included in the β phase of text, Ælfric’s second stage of editing and rearranging, the manuscripts themselves appear on opposite sides of this stemma. Furthermore, according to Clemoes, other parts of Bodley 340+342 are more associated with the α phase of text, while CCCC 162 contains no items associated with this phase.9 Following this text-based assessment, however, several other scholars have concluded the two manuscripts are rather more closely linked than this: Malcolm Godden, the editor of Ælfric’s Second Series of homilies, showed that for certain textual items the readings from Bodley 340+342 and CCCC 162 are exceptionally similar;10 in an article on the later readers of Ælfric, Treharne claims that the texts of Bodley 340+342 are at least partially derived directly from the texts of CCCC 162;11 and Donald Scragg notes the existence of a scholarly assumption that CCCC 162 is merely a copy of Bodley 340+342, then argues against this, claiming that although the two manuscripts share many items in common, they are arranged quite dissimilarly. He concedes, however, that the former manuscript is at least partially a copy of the latter, noting that certain items in the two manuscripts are textually very close.12 Conversely, Lowe disagrees with Scragg’s assessment, and through a detailed investigation of the language in the manuscripts determines that the two are more likely sister manuscripts rather than an exemplar/descendant pair.13 There are therefore firmly established links between these two manuscripts, although scholarly opinion is divided on whether or not CCCC 162 in whole or in part is copied from Bodley 340+342 or vice versa. In this article, I use a combination of the approaches mentioned above in order to refine our understanding of this relationship between these two manuscripts. The first approach I take is a brief analysis and comparison of some of the more unusual texts found in the two; the second is an analysis of the decoration found in the manuscripts. Before moving on to the evidence connecting them, I include here a short description of each of the manuscripts; for more thorough descriptions of the individual volumes see Clemoes,14 Kenneth Sisam’s series of three articles on Bodley 340+342,15 and Scragg’s detailed discussion of the texts in CCCC 162.16

Bodley 340+342 are a set of two individually bound manuscripts, generally considered to be parts of a single compilation due to their complementary contents: together, the two manuscripts contain texts that cover the entire church year, from Christmas to Advent.17 In addition to the textual evidence, both volumes are written in the same main hand,18 and the posited time and location of production is the same for each: most likely Rochester, at the beginning of the eleventh century. Sisam notes that concrete evidence points only to the manuscripts being at Rochester very soon after their creation, and that there is no evidence to exclude outright the possibility that the manuscripts were written at Canterbury and later transferred to Rochester;19 however, later scholarship has often assumed the creation of the manuscripts to have been at Rochester because of this lack of contradictory evidence. Bodley 340+342 contain primarily Ælfrician texts: Bodley 340 contains thirty-one texts, and Bodley 342 contains forty-one; of these seventy-two texts, all but ten are Ælfrician compositions. The non-Ælfrician texts in the manuscripts are anonymous compositions for various Sundays in Lent; an anonymous life of St Paulinus of Rochester (the inclusion of which provides evidence towards this homiliary being created at Rochester); and texts for the three days preceding Easter, on which Ælfric believed that no preaching should be conducted. This was an idiosyncrasy apparently specific to Ælfric, and I return to this presently for further discussion.20

CCCC 162 is a single volume, written in one main hand, likely also produced in either the late-tenth or early-eleventh century, in south-eastern England.21 The manuscript is bound along with leaves from another Ælfrician manuscript (CCCC 178) as well as a woodblock print placed at the beginning of the homiliary; both additions were made in the modern era by Matthew Parker.22 As these sections were not present at the time of the manuscript’s creation, they are not included in the following discussion. The texts in this manuscript cover holidays and feast days for the church year beginning at the second Sunday after Epiphany, and ending at the second Sunday in Advent.23 CCCC 162 contains fifty-eight items, all but thirteen of which are Ælfrician compositions. As in Bodley 340+342, the non-Ælfrician compositions include texts for the three days preceding Easter; in addition to these, the manuscript contains anonymous homilies for various days in Rogationtide and ends with an imperfect text on St Augustine of Canterbury.24 Unlike Bodley 340+342, however, which contain only a single homily for any given day, the compiler of CCCC 162 appears to have included multiple versions of homilies for specific days whenever possible. In Scragg’s chapter on the manuscript, he presents evidence that the manuscript was compiled in at least three independent blocks of writing, the third and largest block containing all the texts referred to in this article.25

Textual Similarities

Twenty-eight texts appear in both CCCC 162 and Bodley 340+342,26 referred to here by their short titles as assigned by Bruce Mitchell, Christopher Ball and Angus Cameron.27 Many of the texts that appear in each of the manuscripts occur frequently elsewhere: for example, while Ælfric’s homily for Pentecost28 appears in each of the homiliaries, it is also the most frequently copied text from both series of Ælfric’s Catholic Homilies overall, surviving at least in part in sixteen different manuscripts. In addition to ÆCHom I, 22, the homiliaries both contain many of Ælfric’s homilies for the Sundays following Pentecost, often in a similar or sometimes identical order. However, as these texts are very common, neither their appearance in these manuscripts nor the order in which they appear is necessarily significant. Furthermore, as many of the homilies are titled referencing their purpose for preaching on specific Sundays following Pentecost, placing the homilies in the sequence in which they appear here was not necessarily an active decision made by the compiler as much as it was simply following the church calendar and order as suggested by the texts.

More unusual, and therefore more enlightening, are those texts in the homiliaries that are not attributed to Ælfric. Both Bodley 340+342 and CCCC 162 contain several non-Ælfrician, anonymous texts, and both manuscripts in fact contain four of the same anonymous homilies. In both, these texts are the four homilies that appear directly before Ælfric’s Easter-related homilies (here defined as Ælfric’s Easter homily, Ælfric’s Palm Sunday homily, and Ælfric’s homilies for the two Sundays following Easter). The inclusion of these anonymous texts in both Bodley 340+342 and CCCC 162 is likely due to Ælfric’s aforementioned belief concerning the days before Easter: he referred to these days as swig-dagas, ‘silent days’,29 and never composed or included homilies intended to be preached on these days in the manuscripts that he had supervision over. Ælfric’s reasons for this belief are unknown, and based on the surviving evidence, it appears to have been unique: in almost all manuscripts containing Ælfric’s texts for Easter, compilers have supplemented the Ælfrician collection with anonymous homilies for the days preceding Easter itself, against Ælfric’s explicit requests found in his English preface not to include additional texts alongside his own.30

Both of the homiliaries discussed in this paper fill this gap with identical selections of anonymous homilies. In CCCC 162 and Bodley 340+342, the anonymous texts that appear in both sets of manuscripts are as follows: an anonymous homily referred to as In cena domini (HomS 22), an anonymous homily intended for preaching on Good Friday (HomS 24 In parasceve), an anonymous homily on Holy Saturday (HomS 25 In sabbato sancto), and an anonymous homily on Palm Sunday (HomS 18 In dominica palmarum). These are not especially common homilies, but they are found elsewhere; HomS 22 appears in three other manuscripts: Cambridge, Corpus Christi College, MS 198; Cambridge, Corpus Christi College, MS 302; and British Library, Cotton Faustina A. ix; according to Lowe, further anonymous homilies for this day survive in an additional two manuscripts.31 HomS 24 appears in two other Ælfrician manuscripts, as well as in the Vercelli Book, a manuscript completely unrelated to the Ælfrician textual tradition;32 HomS 25 appears in only one other manuscript; as does this version of a homily for Palm Sunday (HomS 18). While these anonymous vernacular homilies do exist in multiple manuscripts, the number of manuscripts in which they occur is still low compared to the frequency with which Ælfrician texts survive. That all four texts exist in the same two homiliaries is thus noteworthy. Furthermore, the compilers of both CCCC 162 and Bodley 340+342 very likely had access to multiple vernacular homilies to select from for inclusion for these dates: two other vernacular versions of a homily similar to HomS 22 survive in other manuscripts, and CCCC 162 alone contains two anonymous, vernacular versions of a homily for Palm Sunday, in addition to the Ælfrician one on pp. 305–22. Therefore, not only did the compilers of Bodley 340+342 and CCCC 162 both choose to include non-Ælfrician homilies, they chose to include the very same ones in lieu of any others that may have been available.

The sections of Bodley 340 and CCCC 162 that contain the homily for Easter and its surrounding holy days are highly similar. In both, the sequence of texts runs HomS 18, HomS 22, HomS 24, and HomS 25, followed in Bodley 340 by ÆCHom I, 15, and another anonymous Palm Sunday homily, HomS 27, in CCCC 162. Both manuscripts then converge once again with ÆCHom I, 16 and ÆCHom I, 17. Furthermore, based on textual evidence in CCCC 162, it is likely that these sections were originally intended by their compilers to be identical. As it stands, the two passages are identical, save CCCC 162’s inclusion of a second anonymous Palm Sunday homily in place of an Easter homily. However, this difference likely does not represent a deviation in the intended order of texts between the two manuscripts, but rather an instance of scribal error: the text following the Holy Saturday homily was most probably intended to be Ælfric’s Easter homily. There are several pieces of evidence that indicate that the inclusion here of the second Palm Sunday homily was an error: first, and perhaps most convincingly, the rubricator of CCCC 162 rubricated the second Palm Sunday homily in CCCC 162 as In die sancto pasce, either assuming that a text on Easter Sunday would logically be included here, or perhaps working from an exemplar that did place an Easter Sunday text in that position. Furthermore, the inclusion of an additional Palm Sunday homily following a homily on Holy Saturday makes no sense when viewed in the context of the rest of the manuscript: the date of Palm Sunday, as a moveable feast in the church year, is without fail one week before Easter. The rest of the manuscript is laid out in an orderly fashion in accordance with the church calendar; a deviation of this level would be a highly unusual choice on the part of the compiler. Finally, in every other manuscript containing Ælfrician texts that also includes a text on Palm Sunday, Ælfric’s homily on Easter, ÆCHom I, 15, either follows this Palm Sunday homily immediately, or is separated from the homily only by texts for the days falling between the two Sundays in the church year (Good Friday, Holy Saturday, etc.). There is no convincing reason to believe that in CCCC 162, the compiler made a decision that differs so dramatically from both all other surviving compilations and the logical progression of the church calendar. Taking these pieces of evidence into account, we must believe that this sequence, as planned, was intended to have been identical in Bodley 340+342 and CCCC 162. Therefore it is likely that both manuscripts aimed to produce the following set of texts: HomS 18; HomS 22; HomS 24; HomS 25; ÆCHom I, 15; ÆCHom I, 16; and ÆCHom I, 17. Because of either compiler or scribal error, this sequence was not achieved in both.

Decorative Evidence

It is well established that certain aspects of the physical appearance of manuscripts were copied alongside texts, including multiple examples of especially elaborate and carefully decorated initials. Such examples include an almost-identically constructed H in two manuscripts created at the nearby centers of Christ Church and St Augustine’s;33 foliate designs associated with the beginnings of certain texts, e.g. the large B and following display script beginning Psalm I in several manuscripts;34 and complex ruling patterns, especially when copying biblical texts with many layers of marginal commentary meant to accompany them.35 However, neither Bodley 340+342 nor CCCC 162 are especially luxurious manuscripts, as is to be expected from vernacular homiliaries from this period: according to Gameson, ‘Anglo-Saxon reading books and homiliaries — the majority of the extant material in Old English — were not by and large decorated’.36 That said, despite this overall dearth of decoration in vernacular manuscripts, decorated initials were ‘within the reach of almost any scriptorium’,37 and are found in each of these two homiliaries, generally to signify textual boundaries or changes in language used in the body of the text.

There are many common elements between the two manuscripts: enlarged initials, the use of colored inks, and decorated letters indicating the beginnings of individual texts appear throughout each. In both, the vast majority of the decorative elements are concerned with the demarcation of individual texts, more specifically used to signal clearly to the reader that a new text has begun. Individual texts in these manuscripts follow a somewhat standard layout for an Old English prose text in this period: they begin on a new line, generally with a Latin title or rubric written in red ink. Following this, many of the texts contain a short Latin incipit relating to or introducing the Old English text, written in dark brown/black ink, and sometimes written in a different style of script from the Old English. The first letter of this incipit is often written in colored ink and enlarged slightly in comparison to other majuscule letters in the text. Following this is the beginning of the Old English text, the body of which is also written in dark brown ink. The first letter, or letters, of this text may be enlarged, written in colored ink, or elaborated upon with decorative features. Only very rarely in Bodley 340+342 are these text-opening initial letters decorated with zoomorphic forms; no zoomorphic decorations are present in CCCC 162.

To analyze the decorations found in these manuscripts, I use a set of criteria that allows me to discuss aspects of decoration in a quantifiable way, specifically with regards to enlarged initials. This method of scoring concerns only enlarged initials, and not any other aspects of decoration or quality of the manuscript. This ‘score’ is therefore reflective of the incidence of enlarged or decorated initials only, and is more illustrative in manuscripts with a larger number of decorated letters: a manuscript with only two or three highly decorated initials may appear from its ‘score’ to be a more decorated manuscript overall than one with a great deal more decorated letters, each of which is less elaborate. The decoration level should therefore be seen as a tool indicating areas of interest and further investigation, rather than an end in and of itself. Both manuscripts discussed here have a fairly high number of decorated initials, 102 in Bodley 340+342 and 93 in CCCC 162, making a cross-manuscript comparison of this type appropriate. This methodology has been developed following on from Johanna Green’s work on the Exeter Book,38 in which she uses the size and distribution of litterae notabiliores, among other textual evidence, to argue that Judgement Day I, Resignation A, and Resignation B were intended as a single conceptual unit within the manuscript. The work is also indebted to Murray J. Evans’ monograph that uses statistical analysis to assess the implications of genre on manuscript decoration: in the monograph, Evans concludes using this analysis that texts described generically as romances are indeed more decorated overall than texts of other genres; he further determines that this conclusion holds true regardless of where the text appears in the manuscript.39 In addition to these two pieces of work, there have been a number of other studies using decorative — and importantly, non-illustrative — information to determine manuscript relations: a similar methodology developed by Ruth Carroll and others has been used to analyze the physical appearance of the manuscript page in a way that allows for the ‘visual pragmatics’ of the page to be read, defined as ‘anything on the page that adds meaning to the linguistic message’;40 Keskiaho has used written marginalia appearing in copies of Augustine’s De Civitate Dei to refine manuscript stemmata;41 and an illustrated border in a nineteenth-century manuscript containing Njáls saga has been identified as originating from a 1772 edition.42 What differentiates the methodology used for the current study is its combination of qualitative measurements as well as its application in further creation of Ælfrician manuscript stemmata: this study allows the transmission of texts and manuscripts to be tracked in a way not otherwise possible in generally undecorated homiliaries. I determined the most relevant aspects in Ælfrician manuscripts to be the size, color, and level of decoration of each of the decorated letters in Bodley 340+342 and CCCC 162. The size of each initial was measured in line height, and the color was noted (red, green, a combination of two, etc.); these aspects of quantification are straightforward and thus easy to record and compare. The level of decoration of a letter, however, is more abstract and thus more difficult to quantify with a single descriptive word. In order to facilitate a straightforward analysis of the data, I created a set of five categories, with the numbers 1–5 describing the amount of elaboration on each initial. The table below contains a further description of what each of these numbers represent. These numbers are assigned to the letters that begin texts, as well as any letters throughout the entire manuscript that are enlarged, written in colored ink, or any combination of the above characteristics. It should be noted that all levels of decoration are determined within the Ælfrician First Series manuscript corpus alone, and not compared to other surviving manuscripts written in Old English: a letter that is quite heavily elaborated in an Ælfrician manuscript would likely appear somewhat lacklustre in comparison with one from a more elaborate Old English manuscript.

Table 1: categories of decoration.
Numerical value Description of letter
1 The letter is not decorated at all and is written in much the same style as an average majuscule letter in the body of the text.
2 The letter is very slightly decorated: a single foliate sprig, or a single dot along a pen-stroke, for example.
3 The letter is somewhat decorated: more elaborate than a letter with a value of 2 (but not especially so), and not exhibiting the more intricate features of a letter with a value of 4.
4 The letter is very heavily decorated, often with multiple colors of ink. Despite its high level of decoration, the letter is neither illuminated nor elaborated upon with zoomorphic forms.
5 The letter incorporates zoomorphic motifs and is thus the rarest and most highly decorated type of letter in the manuscripts.

By assigning these values to every enlarged, decorated, or colored initial in each of the manuscripts discussed here, individual initials can be compared with a degree of specificity. In addition to this, an average overall ‘decoration level’ of the initials from each manuscript may be determined by finding the average of the values of each decorated initial. This resulting decoration level then gives an at-a-glance impression of the extent to which the manuscript is decorated in comparison to the other manuscripts in the corpus. All Ælfrician First Series manuscripts were assessed in this way; several additional vernacular manuscripts from the period were also scored. My recently submitted doctoral thesis considers the full set of First Series manuscripts in this way, and provides more information and context regarding these ‘scores’.

Bodley 340+342 is a relatively heavily decorated homiliary, and the two volumes make up the second and third most decorated manuscripts in the corpus. As Old English homiliaries were only rarely decorated,43 heavily decorated text-beginning letters (especially zoomorphic ones) are extremely uncommon within these manuscripts. However, between the two volumes of this homiliary, seven letters are elaborated upon with zoomorphic motifs. Only one of the seven initials are to be found in Bodley 340 and it appears at the very beginning of the manuscript;44 all of the remaining initials with a decoration value of 5 occur in Bodley 342, perhaps indicating that different decorators were available during the production of the two volumes. This supposition is complicated by the fact that a single main hand wrote the majority of the two;45 it is possible, then, that whoever supervised the manuscript’s creation may have requested more elaboration in the second portion of the homiliary. The zoomorphic initials are of several different types: some, like those decorating the beginning of each volume, use beasts’ heads and terminal foliate designs to elaborate upon the basic shape of the letter. In others, the form of the letter in some is composed entirely of animals, such as the S on f. 21r, which is made up of two winged beasts attached at the tail. These distinctively decorated letters, along with the other highly decorated letters in the manuscript, give Bodley 340+342 an overall decoration level of 2.28, making it the second most decorated homiliary amongst all manuscripts containing Ælfric’s First Series of Catholic Homilies. The only First Series manuscript with a higher decoration level than Bodley 340+342 is London, British Library, Cotton Vespasian D. xiv, ff. 4–169, an especially late manuscript, produced s. xiimed, that contains a high percentage of non-Ælfrician texts.46

CCCC 162 does not contain any zoomorphic illustrations similar to those found in Bodley 340+342. However, the manuscript does contain a high number of initials that are relatively intricately decorated, with many using foliate designs, dots, and fleurs-de-lys to elaborate upon the basic letter-forms. The level of decoration found in most of the text-beginning initials in this manuscript is unusual within the Ælfrician First Series corpus: of the 93 enlarged or decorated letters within the manuscript, 44 have a decoration level of 3 or 4. Because of the high number of finely decorated letters included within it, CCCC 162 has an overall decoration level of 2.2, only very slightly less than that of Bodley 340+342, making it the third most decorated manuscript within the Ælfrician First Series corpus.

In addition to this broad approach, a closer investigation into certain individual letters in the manuscripts reveals further similarities. As mentioned above, ÆCHom I, 22 is one of Ælfric’s most frequently surviving works, and appears in part in sixteen manuscripts. Because of the text’s wide distribution and high level of occurrence in surviving manuscripts, it is possible to conduct a comparative analysis on this single homily. For the purposes of the present study, the subject of focus is the letter that begins the Old English text of the homily, which is F in all instances of the text (the body of the text begins with some variation of Fram þam halgan easterlican dæge in all manuscripts). The Fs that begin this homily vary quite widely across all the texts, from as short as 2.5 lines in height to as tall as 8; and from decoration levels ranging from 1 to 5. The Fs that begin ÆCHom I, 22 in Bodley 340+342 and CCCC 162, however, share several key similarities not found amongst the other initials in the corpus.

In Bodley 342, ÆCHom I, 22 begins at the top of the first page of a new quire on f. 57r. This F is very large in comparison to the other initials in the manuscript: the letter is 7.5 lines tall, while the average size of enlarged or decorated letters across the entire manuscript is only 2.9 lines tall. For further context, the only letter in the entire manuscript larger than this is the A that opens the manuscript, which is ten lines tall; the next largest letter after the F opening ÆCHom I, 22 is another F on f. 147v, this one only six lines tall. In addition to its unusual size, this F on f. 57r is also the most elaborately decorated letter in the entire manuscript, as well as perhaps in the entire corpus of Ælfrician manuscripts as a whole. This F is most similar to one of Wormald’s Type I initials: ‘a letter constructed, at least partially, from complete or near-complete representations of animals and birds’.47 Despite the frequency of letters of this type from the first half of the tenth century through the late Anglo-Saxon period, very few other initials of this type exist in the corpus. The body of this F is composed of a line drawing done in black ink, beginning with the head and body of a bird-like beast, with wings and a beak; the tail of the beast is extended several lines down into the left-hand margin, and ends with a flourish. The lower arm of the F is created by another winged beast, more dragon-like than bird-like, biting the tail of the bird-like beast. The dragon-like beast’s body also ends with penwork flourishes.

In CCCC 162, ÆCHom I, 22 begins mid-quire on p. 441. ÆCHom I, 22 does not begin at the top of a new page in this manuscript: the rubric of the homily occurs in the third line, and the enlarged initial and text itself begins on the fourth line of the page. Despite this difference, the text-beginning initial is exceptionally large in this manuscript as well, at 8.5 lines tall. The only larger letter in this manuscript is, as in Bodley 342, the text-beginning A that opens the first text in the manuscript on p. 1, at nine lines in height. The F beginning ÆCHom I, 22 is not particularly intricately decorated in comparison to some of the other initials in the manuscript, with a decoration level of only 3, but its unusual size marks it as different from the other text-beginning initials in the manuscript. The letter is written in green ink, with the body of the letter split down the middle by a strip where parchment has been allowed to show through; this split continues along the arms of the F, giving the impression of a smaller F made of negative space within the larger green one. The ends of the arms and body of the F are finished with flourishes, also in green; as with the F in Bodley 342, a decorative descender on the bottom of the letter extends far down the left-hand margin of the page.

This decorative descender found in both in CCCC 162 and Bodley 340+342 merits further discussion. It is by no means a standard feature of the Fs within the corpus: in all of the versions of this homily, only three begin with an F that could be described as having an element of decoration that descends into the left-hand margin of the page. The third manuscript that has a descender of this sort is Oxford, Bodleian Library, Bodley 343, another late manuscript posited to have been created in the second half of the twelfth century.48 Of the remaining thirteen manuscripts that contain all or part of ÆCHom I, 22, two of them are incomplete or lacking the text-beginning letter completely,49 and the other eleven have standard or decorated serifs at the bottom of the letters rather than a flourish that descends into the left-hand margin. Even in another manuscript that contains many of the same anonymous texts as these two, Cambridge, Corpus Christi College, MS 198, the F beginning ÆCHom I, 22 is very different from these two: the F in CCCC 198 is only four lines tall (a tie with seven other enlarged initials for the undistinguished spot of joint-fifth-tallest initial height in the manuscript), with no decoration at all aside from a very thin line of negative space down the body of the F itself. Unlike the Fs in Bodley 342 and CCCC 162, which are unusual in their size, decoration, or both, the F in CCCC 198 is neither one of the largest nor one of the most decorated letters in the manuscript. Indeed, this appears to be the case for most of the other Fs: while Pentecost was clearly an important date for preaching (as indicated by the text’s frequent inclusion within these homiliaries), none of the other Fs that begin the homily occupy such a unique position in the manuscript as the two in Bodley 342 and CCCC 162.

A further similarity between the two Fs is their appearance in relation to the other text-beginning letters in the two manuscripts. That the initials on these pages in Bodley 340+342 and CCCC 162 are larger than all others is to be expected, considering that the first pages of manuscripts are frequently the most intricately and elaborately decorated. What is more surprising is that ÆCHom I, 22 has been given almost equal prominence in both Bodley 342 and CCCC 162, despite this prominence not occurring in any of the other manuscripts that contain this text. In Bodley 342, ÆCHom I, 22 starts at the beginning of a new quire, which could perhaps partly explain the emphasis paid to the letter: it is conceivable that whoever was in charge of decorating the start of a new quire may have been struck with additional vigour for the task. However, the F in CCCC 162 does not begin at the start of a quire, but rather in the middle of a stint of writing. Someone involved in the production of CCCC 162 clearly considered the Pentecost homily significant, giving it an uncommonly large opening-initial, despite its otherwise unremarkable nature in both its content and positioning within the manuscript.

There are two possible interpretations, then, of the motivation behind the creation of this highly decorated F in CCCC 162. The first possibility is that it was simply happenstance: during the long process of decorating a nearly-complete manuscript, an artist decided that a decorative descender was appropriate for this enlarged initial, and the style and size of F is only coincidentally similar to another F from another manuscript; both decorators independently decided to place similar visual importance on the two texts in a manner that is otherwise unique in all other instances of the homily in the Ælfrician corpus. The alternative, and perhaps the more likely, interpretation is that these two manuscripts are very closely related in a manner revealed by the similarities between the two Fs. As suggested by Scragg, despite the thoughtful and original compilation of CCCC 162, parts of the manuscript were indeed likely copied from Bodley 340+342. In addition to the textual evidence for this relationship discussed by Scragg, Clemoes, Godden, and others, this similarity found in the two ÆCHom I, 22 Fs adds another type of evidence: not only that the two manuscripts are indeed related, but that the scribe responsible for the decorated initials in CCCC 162 was directly inspired by the visual cues set by the scribe of Bodley 340+342. This would explain both the high level of similarity in the texts found between the two manuscripts, as well as the appearance of the F in CCCC 162. Of course, there are complications to this claim: as very often multiple manuscripts were used as exemplars when creating a new manuscript, Bodley 340+342 and CCCC 162 share connections with many other manuscripts as well as with each other. It is also possible that rather than one being a direct copy of the other, they both stem from a single lost exemplar that contained a similarly exceptional F, and it is mere coincidence that ÆCHom I, 22 begins at the start of a new quire in Bodley 342; this theory would therefore adhere more closely to Lowe’s posited stemma.50 This would also perhaps indicate that no other surviving manuscripts were copied from this lost exemplar, or all other daughter manuscripts disregarded this enlarged F. Regardless, whether the manuscripts are an exemplar/descendant pair or sisters, they appear to be very closely related. The similarity between these two Fs of course does not necessarily provide definitive proof for one posited stemma over the other; however, the similarity of the F in each provides a new detail within the body of evidence concerning how the two volumes may be related.

Conclusion

In this article, I have described how combining two often disparate approaches to manuscript analysis can provide us with a new angle of inquiry, even towards manuscripts for which a great body of scholarly work already exists. Through a combination of textual comparison and an analysis of decoration, I have noted a combination of textual and visual similarities between Bodley 340+342 and CCCC 162 that do not exist in any other Ælfrician First Series manuscripts, supplying an additional piece of evidence for the argument that the two manuscripts share a particularly strong link. The scribes that worked on these relatively simple manuscripts did not fundamentally change the way in which they copied texts and produced manuscripts when creating these volumes: the scribes and rubricators took visual inspiration when copying even these less-elaborate initials, and carried exceptional letters from manuscript to manuscript along with the texts. By investigating these simpler text-beginning initials in the same manner one would treat a more elaborate decorative composition, we can construct a complementary relational stemma that augments those created through more traditional methods. This is especially useful in cases such as this, where despite the centuries of scholarly interest in Ælfric and his works, there are still many unknowns concerning how the manuscripts were created and what networks of scriptoria they moved through. Even though these vernacular homiliaries may be less visually impressive than other manuscripts from the period, it is well worth our time to investigate the decorative features they do contain, in Ælfrician manuscripts as well as others: this approach may be applied to any codex with any level of extra-textual elaboration, no matter how minimal. Rather than treating these less-decorated manuscripts as simply physical objects from which important texts are to be extracted, a closer look at the decorative aspects that do exist can reveal intriguing similarities that may otherwise have gone unnoticed.

1 Elaine M. Treharne, ‘The Production and Script of Manuscripts Containing English Religious Texts in the First Half of the Twelfth Century’, in Rewriting Old English in the Twelfth Century, ed. by Mary Swan and Elaine M. Treharne (Cambridge: Cambridge University Press, 2000), pp. 11–40; Elaine M. Treharne, ‘Making Their Presence Felt: Readers of Ælfric, c. 1050-1350’, in A Companion to Ælfric, ed. by Hugh Magennis and Mary Swan (Leiden: Brill, 2009), pp. 399–422.

2 The Production and Use of English Manuscripts 1060 to 1220, ed. by Orietta Da Rold and others (Leicester: University of Leicester, 2010), <https://www.le.ac.uk/english/em1060to1220/index.html>.

3 Ælfric’s Catholic Homilies: The First Series, ed. by Peter Clemoes (Oxford: Oxford University Press, 1997).

4 Clemoes, pp. xvii–ii; it is worth noting here that Clemoes’ and Godden’s work does not account for the entire transmission and history of Ælfric’s texts; for a more recent and in-depth discussion of the Ælfrician canon see Aaron J. Kleist, The Chronology and Canon of Ælfric of Eynsham (Cambridge: Boydell & Brewer, 2019); and for a discussion of Ælfrician works in other languages see John Frankis, From Old English to Old Norse: A Study of Old English Texts Translated into Old Norse with an Edition of the English and Norse Versions of Ælfric’s De Falsis Diis (Oxford: The Society for the Study of Medieval Languages and Literature, 2016).

5 Clemoes, pp. 64–97; 134–68.

6 Clemoes, p. 160. This set of manuscripts is considered by Clemoes to be a single collection, based on the use of a single hand across all three manuscripts and the complementary contents of the three volumes p. 41.

7 Clemoes, p. 154.

8 Kathryn A. Lowe, ‘Filling the Silence: Shared Content in Four Related Manuscripts of Ælfric’s Catholic Homilies’, Digital Philology: A Journal of Medieval Cultures, 4.2 (2015), 190–224 (p. 199).

9 Clemoes, pp. 137–44.

10 Æelfric’s Catholic Homilies: The Second Series, ed. by Malcolm Godden (Oxford: Oxford University Press, 1979), p. xxii.

11 Treharne, ‘Making Their Presence Felt: Readers of Ælfric, c. 1050-1350’, p. 408.

12 D. G. Scragg, ‘Cambridge, Corpus Christi College 162’, in Studies on Anglo-Saxon Manuscripts and Their Heritage, ed. by Phillip Pulsiano and Elaine M. Treharne (Aldershot; Brookfield, Vt: Ashgate, 1998), pp. 71–84 (p. 78).

13 Lowe, pp. 199–200.

14 Clemoes, pp. 7–10, 13–6.

15 Kenneth Sisam, ‘MSS. Bodley 340 and 342: Ælfric’s Catholic Homilies’, The Review of English Studies, 7.25 (1931), 7–22; Kenneth Sisam, ‘MSS. Bodley 340 and 342: Ælfric’s Catholic Homilies’, The Review of English Studies, 8.29 (1932), 51–68; Kenneth Sisam, ‘MSS. Bodley 340 and 342: Ælfric’s Catholic Homilies’, The Review of English Studies, 9.33 (1933), 1–12.

16 Scragg.

17 Clemoes, p. 8.

18 Sisam, ‘MSS. Bodley 340 and 342 (1931)’, p. 10.

19 Sisam, ‘MSS. Bodley 340 and 342 (1931)’, p. 11.

20 Joyce Hill, ‘Ælfric’s “Silent Days”’, Leeds Studies in English, 16 (1985), 118–31.

21 Clemoes, p. 13.

22 ‘Cambridge, Corpus Christi College, MS 162: Old English Homilies’, Parker Library On the Web: Manuscripts in the Parker Library at Corpus Christi College, Cambridge <https://parker.stanford.edu/parker/catalog/ft757ht3699>.

23 Clemoes, pp. 13–14.

24 Clemoes, pp. 13–14.

25 Scragg, p. 76.

26 The Ælfrician texts that occur in both CCCC 162 and Bodley 340+342 are as follows: ÆCHom I, 1; ÆCHom I, 8; ÆCHom I, 10; ÆCHom I, 11; ÆCHom I, 16; ÆCHom I, 17; ÆCHom I, 19; ÆCHom I, 20; ÆCHom I, 22; ÆCHom I, 23; ÆCHom I, 24; ÆCHom I, 28; ÆCHom I, 33; ÆCHom I, 35; ÆCHom I, 39; ÆCHom I, 40; ÆCHom II, 5; ÆCHom II, 6; ÆCHom II, 19; ÆCHom II, 24; ÆCHom II, 26; ÆCHom II, 27; ÆCHom II, 28; ÆCHom II, 31. In addition to these, each manuscript also contains the same four anonymous homilies mentioned in the body of the text.

27 Bruce Mitchell, Christopher Ball, and Angus Cameron, ‘Short Titles of Old English Texts’, Anglo-Saxon England, 4 (1975), 207–21.

28 ÆCHom I, 22.

29 Hill, p. 118.

30 Lowe, p. 191.

31 Lowe, p. 220.

32 Lowe, p. 200.

33 Richard Gameson, ‘Book Decoration in England, c. 871–c. 1100’, in The Cambridge History of the Book in Britain, ed. by Richard Gameson (Cambridge: Cambridge University Press, 2011), i, 249–93 (p. 261).

34 Richard Gameson, ‘English Manuscript Art in the Mid-Eleventh Century: The Decorative Tradition’, The Antiquaries Journal, 71 (1991), 64–122 (p. 81).

35 Malcolm Beckwith Parkes, ‘Layout and Presentation of the Text’, in The Cambridge History of the Book in Britain, ed. by Richard Gameson, 5 vols (Cambridge: Cambridge University Press, 2011), i, 55–74 (p. 61).

36 Gameson, ‘Book Decoration in England’, p. 287.

37 Gameson, ‘Book Decoration in England’, p. 287.

38 ‘Judgement Day I, Resignation A and Resignation B: A Conceptual Unit in the Exeter Book’ (unpublished doctoral thesis, University of Glasgow, 2012).

39 Rereading Middle English Romance Manuscript Layout, Decoration, and the Rhetoric of Composite Structure (Montreal: McGill-Queen’s University Press, 1995).

40 ‘Pragmatics on the Page’, European Journal of English Studies, 17.1 (2013), 54–71 (p. 56).

41 Keskiaho, Jesse, ‘Copied Marginal Annotations and the Early History of Augustine’s “De Civitate Dei”’, Augustiniana, 2, 2019, 277–98.

42 Svanhildur Óskarsdóttir, ‘Introduction’, in New Studies in the Manuscript Tradition of Njáls Saga: The Historia Mutila of Njála, ed. by Emily Lethbridge and Svanhildur Óskarsdóttir (Kalamazoo, MI: Medieval Institute Publications, 2018), pp. xiii–xxiii (p. xviii).

43 Gameson, ‘Book Decoration in England’, p. 287.

44 The seven initials that are decorated in this manner are as follows: H in Bodley 340 on f. 1r; A in Bodley 342 on f. 1r; S in Bodley 342 on f. 21r; F in Bodley 342 on f. 57r; D in Bodley 342 on f. 110v; D in Bodley 340 on f. 127v; U in Bodley 342 on f. 135r.

45 Sisam, ‘MSS. Bodley 340 and 342 (1931)’, p. 10.

46 Clemoes, p. 16.

47 Richard Gameson, ‘The Decoration of the Tanner Bede’, Anglo-Saxon England, 21 (1992), 115–59 (p. 116).

48 Clemoes, p. 1.

49 The two manuscripts are London, Lambeth Palace 487, where space has been left for an enlarged initial that has not been written in; and Cambridge, Corpus Christi College, MS 188, where the text begins imperfectly on p. 211, with no title or proper beginning of the text.

50 Lowe, p. 200.