The Heroic Age

A Journal of Early Medieval Northwestern Europe

Founded 1998   |   ISSN 1526-1867

Issue 14 (November 2010)  |   Issue Editors: Eileen A. Joy & Andrew Rabin

The 'Scipmen' Scribe and Cambridge, Corpus Christi College 383

Kathryn Powell

University of Cambridge

© 2010 by Kathryn Powell. All rights reserved. This edition copyright © 2010 by The Heroic Age. All rights reserved.

Abstract:  This article reviews the scholarship on MS Cambridge, Corpus Christi College 383 and particularly examines the case for the manuscript's St Paul's, London, origin. Based on a study of annotations, it suggests that the manuscript may have been produced elsewhere for the bishop of London and then modified at St Paul's.

§1.  Cambridge, Corpus Christi College 383 has attracted a fair amount of scholarly interest over the years, and for good reasons. It is one of the few surviving Anglo-Saxon law manuscripts that has no obvious link with Wulfstan; it is one of the earliest English law manuscripts to be comprised (as it was originally copied) only of legal texts, without homiletic or liturgical texts accompanying them; it has a St Paul's, London, provenance, making it a potentially important source of information about the early history of the cathedral community; and, as a post-Conquest manuscript, it provides an important link between Anglo-Saxon and Anglo-Norman legal texts and traditions. It is right, therefore, that it should be much discussed, and unfortunate that it should not be better understood.

§2.  Much of the interest of CCCC 383 rests in one way or another on the activity of a scribe, who adds to the end of the manuscript, copies of the West-Saxon genealogy and a text known as 'Scipmen'. The latter text almost certainly associates the scribe with St Paul's and thereby gives the manuscript a St Paul's provenance. The further assumption of a St Paul's origin for the manuscript is not inconceivable: we have the example of Wulfstan to associate lawmaking with the bishopric of London. For this reason, the manuscript is frequently described as having been written at St Paul's. Few scholars who have studied the manuscript, however, have noted the Scipmen scribe's other activities in it, which include making corrections to some of the main texts. A close study of these corrections could possibly tell us a great deal more about the manuscript's origins. Furthermore, we are fortunate in that Corpus 383 is corrected and annotated by more than one hand, including that of a rubricator who seems to be working shortly after the manuscript was produced. Taken together, his annotations and the Scipmen scribe's have the potential to tell us something about how the manuscript was used at St Paul's, how it was corrected, and potentially what sort of manuscript(s) provided the basis for any corrections. Yet these interventions in the manuscript—particularly the Scipmen scribe's—have not, so far, been investigated fully. While the systematic study this scribe deserves is beyond the scope of a short article, I would like to at least review the scholarship on the manuscript and begin investigating the Scipmen scribe's activity alongside that of the rubricator to see if it can shed any light upon the question of the manuscript's origins.

The Manuscript

§3.  CCCC 383 is a collection of legal texts in Old English, written by a single scribe and traditionally associated with St Paul's, London. It is a small, thin volume, made to be portable, which may suggest that it was conceived as having some use outside of the cathedral archives. The legal texts it contains are varied, including a few major royal codes and pronouncements, more minor codes and treaties, and in some cases sole English copies of such anomalous legal texts as Gerefa. It has been assigned a variety of dates in the eleventh and twelfth centuries, but there are essentially two traditions of dating the manuscript, with one based on Liebermann and the other based on Ker. Liebermann dated the manuscript twice: in his edition of the Scipmen list published in 1900, he merely dated the manuscript to the twelfth century (Liebermann 1900, 23), whereas in his Gesetze, he assigned it a narrow date range of 1125–30, based largely on linguistic evidence (Liebermann 1903–16, 1.xix). This second date is adopted by Robertson in Anglo-Saxon Charters (Robertson 1939, 389), while some more recent legal historians have adopted Liebermann's more general date of twelfth century (e.g., Taylor 2004, 15, n. 115). In his 1957 Catalogue, Ker offered a revised date of 's. xi/xii' on paleographic grounds (Ker 1957, no. 65, pp. 110, 112), and this date has been adopted by many Anglo-Saxonists (e.g., Richards 1986, 181–4; Keynes 1990, 233).1 There are exceptions: Patrick Wormald dates the manuscript to the first quarter of the twelfth century without offering any specific reasons for doing so (Wormald 1999, 228),2 and Susan Kelly adopts Liebermann's linguistic dating of 1125–30, despite citing Ker and Wormald (Kelly 2004, 193). What everyone who has examined the manuscript seems agreed on, however, is that the Scipmen scribe must be working not long after the main scribe of the manuscript. Despite the very different aspects of their hands, they share enough script features to lead one to believe that they are roughly contemporary.

§4.  In general, Ker's dating strikes me as unproblematic. The main hand of the manuscript exhibits a mixture of eleventh- and twelfth-century script features that admittedly makes it quite challenging to date precisely. The assumption of a St Paul's origin for the manuscript is unhelpful in this regard, as little else survives that was definitely written in the vernacular at St Paul's at this time. There are features of the script, however, that lead me to associate it with the late eleventh century or the very beginning of the twelfth. The use of insular r, for example, is suggestive, given that even late eleventh-century scribes begin to use caroline r in vernacular script. G and f are also insular, and the general aspect of the hand shows little of the pointedness of later twelfth-century script. A and s are usually caroline, but these forms are common enough in eleventh-century English vernacular script and do not imply a particularly late date. Furthermore, the manuscript is annotated by a number of hands, two of which—that of the manuscript's rubricator and that of the Scipmen scribe—do not look particularly late. Both hands exhibit a mix of eleventh- and twelfth-century features (although not the same ones in all cases) that suggest that none of the annotators was working later than the early decades of the twelfth century. I therefore have no difficulty in accepting Ker's dating of CCCC 383 as s. xi/xii, and I understand the Scipmen scribe to be working in the early decades of the twelfth century.3

§5.  The main scribe responsible for this collection was quite careless, so that the volume is rife with errors, many of which have been corrected by later hands. Based on the high volume of errors, Patrick Wormald has gone so far as to suggest that the scribe's English was not good. I see little evidence to support such a conclusion, however. Some of the 'errors' Wormald fastens on—faulty word division and inconsistent spelling, for example—are not errors at all, but common practice among eleventh-century scribes copying material quickly and often mechanically.4 Admittedly, CCCC 383 does contain some serious copying errors, but none which indicate anything more than a scribe working quickly and carelessly, just as so many other scribes did throughout the Anglo-Saxon period.5 The proportion and kinds of errors he makes do not differ significantly from, for example, the equally careless work of the two scribes of British Library, Royal 7. C. XII, and no one would suggest that these two did not know English or understand what they were copying. They simply were not paying much attention to the content at the time, perhaps in part because they felt confident that others would review and correct their work.6 The scribe of CCCC 383 likewise makes some corrections to his own work, and a number of later scribes make more, so perhaps he too was copying hurriedly and assuming that he or others could correct his hasty work later. Alternatively, we can know relatively little of the exemplars7 from which the scribe was copying, and it may be that not all of the errors in CCCC 383 originated with him. The simple fact that the manuscript was produced after the Conquest should not lead one to leap to the conclusion that the scribe's English was poor. Recent scholarship has amply demonstrated that knowledge of English did not decline substantially immediately after the Conquest (see, e.g., Swan et al. 2010), and it is even more doubtful that the average scribe's degree of care and accuracy was altered significantly by the events of 1066.

§6.  So, to put it in the best possible light, the main scribe creates opportunities for others to correct and clarify his work, and a number of later readers take advantage of these opportunities. It is notoriously difficult to distinguish between roughly contemporary glossator hands, making it hard to say exactly how many scribes correct or alter the text. Patrick Wormald describes at least two: 'one correcting hand is small, neat, and possibly that of the rubricator, while another one (or two) writes in brown ink and in an angular mode with a notably scratchy quill: he may be identical with the scribe of the diagnostically St Paul's shipmen-list and the West-Saxon genealogy' (Wormald 1999, 234, n. 285). I think it is clear enough that the small, neat hand he describes is indeed that of the rubricator. This hand makes corrections to the manuscript in a dark ink color and habitually uses a crossed 'o' as a mark of revision, as he does on page 1 of the manuscript.8 Most of the rubrics are written in rustic capitals and in red ink, which would make it difficult to compare this annotator's hand with that of the rubricator. However, the main scribe has left no space for rubrics, and the rubricator must fit them in where he can, which means that sometimes he must write them in minuscule characters, and these are easily identifiable with the 'small, neat hand' of the early corrector. Furthermore, the rubricator sometimes makes revisions to the text in red, and uses his characteristic mark of revision in red at the top of page 62.9 So we can track his activity in the manuscript with relative ease as he corrects the main scribe's errors, adds rubrics and, in all likelihood, adds the red textual markers that we see throughout the manuscript.10 His script style is not markedly different from that of the main scribe, and I can see no reason why he might not be making these revisions and corrections very soon after the manuscript was copied. His interventions are clearly aimed at both correcting the manuscript and making it more accessible for readers who do not intend to read it from cover to cover, but are searching it for a particular text or topic. I am less confident about the annotations in brown ink, but suspect they may belong to two scribes whose hands could both be characterized as angular to greater or lesser degrees, the less angular of these belonging to the scribe of the Scipmen list. While I would not claim to be able to distinguish between these latter two in every instance, I think it is possible to point to places in the manuscript where the Scipmen scribe is definitely at work, and those sites of activity can perhaps tell us something about him and his interest in the manuscript.

The Scipmen Scribe and the Scipmen List

§7.  On a blank leaf at the end of CCCC 383, what looks like an early twelfth-century hand has added two texts—a copy of the West-Saxon genealogy and the Scipmen list. The latter is a short but very significant text comprised of a list of contributions required from estates belonging to St Paul's toward the manning of a ship in the English fleet.11 No context or explanation is provided; the text simply begins with the word '[S]cipmen', and there follows a list of place-names and numbers. Apparently, the estates mentioned were expected to pay for the provisioning of sailors, and the numbers specify how many sailors each estate was expected to provision. It is generally agreed that the estates would have been expected to pay a tax, rather than to draft men (Kelly 2004, 193).12 Liebermann dates the text to c. 1000, and his dating is still accepted (Liebermann 1900). Such a date would mean that the text was composed during the latter part of Æthelred the Unready's reign and likely constitutes another witness to the Viking incursions of that period. Depending on how one reads the text, it may suggest that the St Paul's estates, taken together, were expected to provision enough sailors to man a sixty-oar longship—a plausible size for ships of Æthelred's reign.13 The text thus provides an important witness to naval defense in the reign of Æthelred and supplements the evidence of charters and the Domesday Book regarding the land holdings of St Paul's.

§8.  The addition of the Scipmen list to the manuscript has been understood as establishing its St Paul's provenance. The text would be of obvious interest to the St Paul's community and of dubious interest to anyone else, making it most likely that it was copied at St Paul's. In the absence of any evidence to connect the Scipmen scribe with another centre, it seems safe to assume that he was working at St Paul's, which means that the manuscript was there not later than a couple of decades after it was written. Because the Scipmen scribe is generally assumed to have been working not long after the manuscript was produced, it has often been suggested that the manuscript had its origins at St Paul's as well (Wormald 1999, 234; Kelly 2004, 193; Taylor 2004, 12, n. 61). This strikes me as a much more tenuous conjecture. A manuscript would not need decades, or even years, to move from one house to another, and there is every reason why St Paul's, which had a bad habit of burning down, might need to acquire manuscripts copied elsewhere.14 So the manuscript's early twelfth-century St Paul's provenance should not imply a St Paul's origin. If the manuscript was made there, however, one might expect the exemplar or exemplars still to be there when the Scipmen scribe was working. Again, manuscripts do move and one could never be absolutely sure that the exemplar would remain at St Paul's. If it did, however, and was used by later correctors, that would strongly suggest that the manuscript was written at St Paul's. Particularly, if the Scipmen scribe not only adds the Scipmen list and the West-Saxon genealogy to the end of the manuscript but also makes corrections to the law texts, and if there is evidence that he is making those corrections based on an exemplar, then that may go some way toward establishing the manuscript's St Paul's origin. First, however, one must be certain that the Scipmen scribe makes corrections to the manuscript and be able to identify them.

§9.  As Wormald notes, it can be difficult to differentiate the Scipmen scribe from the other scribes active in the margins of the manuscript. When he writes the Scipmen list in a blank space at the end of the manuscript, it is in a large, neat bookhand, while he rarely has so much space in which to work elsewhere in the manuscript. One place where an annotator does have some room is in the margins of page 11, and here I think it may be possible to identify the Scipmen scribe at work.15 At the end of a line of II Athelstan, a later hand has erased a small amount of text and written the words 'Oþer gif' in the right margin, creating a new transition between two sentences. Because he has a relatively wide margin as well as a bit of space at the end of the line over his erasure in which to work, he is able to write in something comparable to a bookhand. The hand shares many of the distinctive characteristics of the Scipmen scribe's hand. Although the ink has bled in the initial majuscule o, one can see that the stroke in the upper left quadrant is thin and flat as in the numerous examples found in the Scipmen-list. The tail of insular g also exhibits the same thin diagonal close stroke that is particularly characteristic of the Scipmen scribe. Throughout the Scipmen list, f is insular with its tongue stroke angled noticeably down to the right and turned up at the end. Where the end of the stroke has worn away or where it is partially overwritten by the following character, the tongue of f may appear to be turned down. If one examines it under magnification, however, one sees that it is invariably turned up at the end. The case is the same with the f of 'gif' on page 11 of the manuscript: the tongue of f slopes notably down to the right and the upturned end of the stroke is almost entirely worn off, so that the tongue appears to be turned down. Under magnification, however, this does not seem to be the case. In short, the form of f is characteristic of the Scipmen scribe's work, even if it does not at first appear so. The ink color and general aspect of the hand match that of the Scipmen scribe's on page 107 of the manuscript,16 and there is little in the brief annotation in the margins of page 11 to suggest that this is not his work.

§10.  Elsewhere, identifying the Scipmen scribe's work is even less straightforward. In addition to the usual difficulties of identifying scribes based on brief, often cramped samples of marginal and interlinear writing, the physical surface of many of the manuscript leaves introduces further troubles. Particularly when one examines the pages under magnification, it is clear that many of them suffered some sort of surface damage before at least some, if not all, of the annotators were working. Most of the damage is due to frequent erasures as part of the annotators' own attempts to correct the many errors in the main text. Even where there are not erasures, however, the surface of some of the pages is very rough, as if heavily rubbed. It is hard to know the cause of this roughness, and it may in fact be a result of how the manuscript parchment was prepared, although it seems to affect the later annotators more than the main scribe. At any rate, the roughness of the surface on some pages appears to affect the adherence of the ink in places. In some annotations, the ink has obviously bled; other annotations appear scratchy where the ink has failed to adhere evenly. I suspect that differences in character width and height, ink color and scratchiness can be misleading in this manuscript, and one has to depend on letter forms for identification much more than one might like, keeping in mind that those letter forms may well be distorted by the confined spaces within which annotators often have to work. Given these challenges, it is not surprising that Wormald found it difficult to judge if there were one or two annotators writing 'in brown ink and in an angular mode with a notably scratchy quill' (Wormald 1999, 234, n. 285). But, given that the Scipmen scribe writes his list in the same shade of brown ink, and since we strongly suspect that he was active in at least one place in the law texts as well, we can begin looking for his activity elsewhere among these angular and apparently scratchy annotations.

§11.  Thankfully, the Scipmen scribe does use several distinctive letter forms, including those already mentioned in connection with the annotation on page 11. Additionally, his s is always long and broken-backed, r is mostly insular but can be caroline in ligature, and round letters (e.g., c, e, o) tend to have an angled south-west quadrant and do not show any trace of a horn. It is hardly possible on that basis alone to positively identify all the Scipmen scribe's annotations, but these letter forms do suggest places in the law texts where he may be at work. We can also easily differentiate him from the rubricator, whose hand is more readily identifiable and who uses crossed o as a mark of revision. We can therefore identify a large number of annotations for which the Scipmen scribe is definitely not responsible as well. Although it is difficult to identify with certainty very many annotations as the Scipmen scribe's, it is certainly possible to point to places where he is probably responsible for corrections or additions, and other places where he is possibly responsible.

§12.  Within these limitations, I think it is still possible to make some interesting observations about the most likely sites of the Scipmen scribe's activity in CCCC 383. First, it is clear that he does not annotate all of the texts in the manuscript. In fact, the annotations which are most clearly identifiable as his are confined to the texts of II Athelstan, Alfred-Ine, and II Cnut, with other possible sites of activity in I Cnut, Forfang, Hundredgemot, and the first of two copies of Alfred-Guthrum in this manuscript. Concentrating for the moment on only the most likely sites of his activity—II Athelstan, Alfred-Ine, and II Cnut—one notices immediately that his most identifiable annotations are concentrated in major royal codes, rather than minor royal pronouncements on specific issues, treaties, and other legal texts. This is interesting in itself, as it may suggest that the Scipmen scribe understood the texts he was modifying and was able to discriminate between different types of Anglo-Saxon law texts. If we examine the texts that he may possibly be annotating, we see that I Cnut fits with his interest in major royal codes, while both Hundredgemot and Alfred-Guthrum would be of interest to Londoners and might catch the attention of an individual who was also interested in the Scipmen list, as all three texts arguably concern London land holdings. If one examines the content of his annotations, I think one can see further evidence that the Scipmen scribe understands the material he is reading and modifying, as some of the changes he makes seem independent of any earlier exemplar. At this point, then, I would like to return to the question of the manuscript's St Paul's origin and ask if the content of the annotations made by the Scipmen scribe, as well as content added by the other annotators, can shed any light on this issue.

The St Paul's Origin of the Manuscript

§13.  As I have mentioned previously, while I see no reason to doubt that the Scipmen scribe was working at St Paul's, the case for the manuscript having been written there strikes me as more speculative. Patrick Wormald makes the case forcefully. He notes that at least some of the corrections in the manuscript seem to be made based on an exemplar, and adds, 'To say that a corrector who may have been the shipmen-list scribe had access to the exemplar(s) is of course to imply that the manuscript itself was written at St Paul's' (Wormald 1999, 234). I suspect that may be putting it a bit too strongly. To say that the Scipmen scribe made some corrections to CCCC 383 based on another manuscript does not necessarily mean that this other manuscript was the exemplar—or even one of a number of exemplars—of CCCC 383. It seems to me that Wormald leaps to this conclusion with the aid of a couple of assumptions. Firstly, he assumes that the main scribe of Corpus 383 does not understand what he is copying—indeed, that he does not even fully understand English. If this is the case, it is indeed unlikely that he would have compiled his manuscript from more than one source, as this would imply understanding the material well enough to make choices about what to include. It is far more likely that an uncomprehending scribe would just slavishly copy a single manuscript, and if there is any indication that the Scipmen scribe makes corrections from that manuscript, then the case for the St Paul's origin of CCCC 383 becomes very strong indeed. If one simply sees the main scribe as a bit careless and hurried but comprehending, however, there is no reason why he or someone directing him might not have chosen to copy this set of legal texts from more than one manuscript source, which complicates the issue. Secondly, although Wormald is clearly aware that the rubricator also makes corrections based on an exemplar, he seems to assume without investigating the matter that the rubricator's corrections are drawn from the same exemplar as the Scipmen scribe's. If the rubricator, who is probably working very shortly after the manuscript was written, is making similar sorts of corrections to the same texts as the Scipmen scribe, then they are probably both working from a manuscript that has always been at St Paul's, and it becomes more likely that CCCC 383 has likewise always been at St Paul's. If both scribes make such corrections throughout the manuscript, then it becomes difficult to doubt that they both have access to the manuscript's exemplar and that CCCC 383 was therefore written at St Paul's. However, one needs to study thoroughly both sets of annotations before drawing such a conclusion.

§14.  When one does examine the manuscript annotations, it is clear that both of these annotators make corrections that must be based on some exemplar. The rubricator in particular frequently corrects instances of eyeskip in the main scribe's work. For example, a hand which appears to be his adds a missed line of text in II Cnut 22.1 (CCCC 383, p. 53, line 17; Liebermann 1903–16, I.324, and n. 4 in apparatus).17 Although the rubricator does a fairly thorough job of correcting this sort of obvious error, the Scipmen scribe still finds a few places to supply text that the main scribe has apparently omitted. For example, at Ine 46.1, the main scribe has copied the clause leaving out the words, 'biþ Wylisc' (CCCC 383, p. 36, line 13; Liebermann 1903–16, I.110, and n. 1 in apparatus). The omission is difficult to pick up, as the resulting statement still makes a kind of grammatical sense: 'Gyf ðonne Englisc mon stalað ga forð ætsace be twyfealdum; gyf hit ðonne onstal ne bið se að na ðe mare'.18 A hand which appears to be the Scipmen scribe's has nonetheless supplied the missing 'biþ Wylisc' above the line between 'ðonne' and 'onstal' in the second part of the clause. Both the difficulty of picking up the error—a difficulty underlined by the rubricator's having missed it—and the occurrence of these words in the closely related Textus Roffensis manuscript suggest that the Scipmen scribe is correcting from an exemplar.19

§15.  It is far from clear, however, that all of the interventions made by these two annotators are corrections based on the exemplar used by the main scribe. The rubricator in particular, working not long after the main scribe, makes a great many corrections that are or could be bringing the texts into line with the exemplar, or at least with some copy of these texts. But he also alters spellings, changes the tense of verbs, adds pronouns that clarify the text but are not strictly necessary to make sense of it, glosses words, and of course adds all the rubrics to relevant places within the text when the main scribe made no provision for them. In short, he makes changes that do not necessarily depend on having an exemplar in front of him. For example, in II As 3.2, where the main scribe has, perhaps uncharacteristically, got it right and written 'swilce' in a place where it both makes sense and agrees with Textus Roffensis, the rubricator has erased the initial s and written an h instead, rendering the clause thus: 7 eac swylc cynges hordere oððe ure gerefena, hwylc ðara ðeofa gewita wære ðe staledon, beo be ðam ilcan'.20 His alteration makes sense, but nothing indicates that he is making it based on an exemplar and, if he is, that exemplar is not obviously related to the source that CCCC 383 and Textus Roffensis share. He may well be acting on his own initiative and altering the text so that it makes better sense to him.

§16.  Interestingly, in at least one place, the rubricator struggles to make sense of an obvious error in a way that strongly suggests that either he does not have access to an exemplar or that his exemplar is flawed. In Ine 28.1, where the text should read 'Gif he ðonne oðierne 7 orige weorðe, ðonne bið he wites scyldig' (Liebermann 1903–16, I.100),21 the main scribe has written 'oðerna' for 'oðierne'—an understandable error, but one that renders the text meaningless. The rubricator has done his best to make sense of the statement by underscoring the -na on the end of oðerna and adding 'do' above the line, which gives us something like, 'If he then should do it again [or poss. do otherwise] and hide. . .' (CCCC 383, p. 32, line 15). Clearly, the rubricator does not have a correct text in front of him and is simply doing what he can to make sense of the passage. The Scipmen scribe's changes, where I can identify them, strike me as even more independent. Many of the errors he corrects are so obvious as to require no exemplar to make them. For example, in I Cnut 22.3, the main scribe has written 'ðeðearf', which makes no sense at all, where he clearly meant 'beðearf' [need], and the Scipmen scribe picks up on it and corrects it—possibly because he has a correct text in front of him, but not necessarily, as it would hardly be difficult to work out on one's own (CCCC 383, p. 46, line 4; Liebermann 1903–16, I.302, and n. 10). Elsewhere, he makes alterations that do not agree with other manuscripts and that seem intended to clarify the text for his audience. For example, in Ine 49.3, in a passage regarding the fine for one's pigs intruding onto another man's pasture, where the manuscript reads, 'Gyf mon nime æbesne on swinum, æt ðryfingrum þæt ðridde, æt twyfingrum þæt feorðe, æt ðumelum þæt fifte', the Scipmen scribe adds the word 'spic' after 'ðryfingrum', apparently assuming an audience that is unaccustomed to being paid in bacon (CCCC 383, p. 37, line 4; Liebermann 1903–16, I.110 and n. 5).22

§17.  In short, both the rubricator, who is working not long after the manuscript was produced, and the Scipmen scribe, who is definitely working at St Paul's, make corrections which must be based on some sort of exemplar, but also make other alterations which seem independent of any exemplar. In various places, they both struggle to make sense of the text in ways that suggest they do not have an exemplar to consult. Furthermore, neither scribe makes interventions throughout the manuscript. In fact, all of their corrections and alterations are confined to the first five quires of the manuscript. They both clearly had a complete manuscript in front of them—the Scipmen scribe's additions are at the end of the manuscript, and the rubricator adds rubrics to texts like Rectitudines and Gerefa that appear later in the manuscript. But I would suggest that they definitely had no other copies of the texts in the later quires available to them for consultation when they were making their corrections. When one examines where they do make corrections and where those corrections seem likely to be based on an exemplar (either because they would have been impossible to make simply based on context or because they agree precisely with more than one other surviving manuscript), they seem to be consulting roughly the same set of texts, probably in the same copy. The rubricator makes interventions in Hundredgemot, I Æthelred, the London version of Alfred-Guthrum, II Athelstan, Alfred-Ine and I and II Cnut, while I am able to identify interventions which are likely made by the Scipmen scribe in Forfang II, the London Alfred-Guthrum, II Athelstan, Alfred-Ine and I and II Cnut. The changes the rubricator makes in Hundredgemot and I Æthelred suggest that he has another copy of these texts in front of him, and both scribes seem to consult another text of Alfred-Guthrum and II Cnut. Interestingly, I cannot identify any alterations made by either scribe to II Athelstan that must be drawn from an exemplar, and both make interventions where they seem only to be guessing at the meaning of the text, so they may not have had another copy of this text to consult. The evidence of I Cnut and Alfred-Ine is mixed. Both scribes make different sorts of corrections here, some of which must be based on an exemplar, and some of which are clearly independent and unique to this manuscript. Both scribes may have had a copy of these texts in front of them, but it may have been substantially different than the one copied by the main scribe of CCCC 383 and/or may have been damaged or incomplete.

§18.  Both scribes' alterations to the London version of A Gu make an interesting case study. CCCC 383 contains two copies of this text. One, found on page 6, is written in the third person and seems to describe the boundaries of the treaty from the perspective of someone situated in London; the other, found on pages 83–4, seems the more official version based on its use of the first person, and describes the boundaries from a less specific perspective.23 Unsurprisingly, all of the alterations made to the London version by these two scribes, as well as those made by other hands, bring this text into agreement with the other, more official version of the text. It is hardly startling that correctors might have realized that the manuscript contained two copies of the same text and chose to correct one against the other. It is the London text they choose to correct, however; no later hand makes any annotations to the official version in this manuscript. Additionally, what the scribes choose not to correct is interesting. While some vocabulary or omitted passages are corrected to match the official version of the text, the third-person point of view that characterizes the London version is retained in contrast with the official version. Likewise, the words that describe the treaty boundaries from the point of view of someone writing on the Thames are preserved, rather than being altered to match the official version. In short, if the annotators are correcting this text against the official version in the same manuscript, they nonetheless all recognize and choose to preserve the integrity of this version as a London text.

§19.  While such a thing is not impossible, and while I do understand both the rubricator and the Scipmen scribe to be reading with interest and understanding the texts that they correct and annotate, I think it unlikely that both of them, along with at least one other scribe who makes alterations to this text, would independently choose to correct the London version against the official version in this careful manner, without even accidently distorting its London character. It seems far more likely, given the pattern of the rest of their corrections, that whatever exemplar they used to make their corrections elsewhere contained another variant copy of the London version, and that all the scribes concerned made corrections with reference to this London copy. Particularly if they were all working in London at St Paul's, it seems entirely plausible that they might have another copy of this version of the text somewhere in their archive, perhaps as part of another legal collection.

§20.  Taken together, then, the evidence of the corrections suggests that while both the rubricator and the Scipmen scribe were likely working in the same place, and while the addition of the Scipmen list and the nature of the corrections to A Gu suggest that this place was probably St Paul's, nothing suggests that the manuscript itself was necessarily written at St Paul's. Both annotators make corrections to the text that must be based on some exemplar, and they both make them in the same parts of the manuscript. In other parts of the manuscript, however, they both seem to struggle to understand and correct the text in ways that suggest they are working without an exemplar in front of them, and much of the manuscript remains uncorrected by either of them. One can only conclude that both annotators had access to a manuscript at St Paul's that contained some—but crucially not all—of the same texts as CCCC 383. If the manuscript was copied from an exemplar or exemplars at St Paul's, then at least one of those exemplars seems to have gone missing soon after the copy was made, before the rubricator made his corrections. Of course, such a thing is not impossible. It is equally possible, however, that CCCC 383 was made elsewhere and given to the London bishop, perhaps to replace a collection of texts lost or damaged in the fire of 1087. It is manifestly not a presentation copy, and one would not expect such material to be compiled in a high-status volume. Rather, it seems intended as a practical reference text, and its small size suggests that it was meant to travel with its owner and to be used outside the confines of the cathedral. Such a practical text might well have been produced hastily and perhaps not to the highest of standards if it was needed urgently, and efforts might have been made at a later date to correct it and make it more readable. To suggest that such a vernacular law text might have been of practical use, let alone urgently required, goes against the conventional wisdom which suggests that the Anglo-Saxon laws were rarely read and not consulted at legal proceedings or meetings of the witan.24 But the physical evidence of Corpus 383 cannot simply be dismissed. At the very least, whatever its origins, it is a portable collection of Anglo-Saxon laws that interested more than one reader enough to inspire them to carefully correct and clarify it. And if it was not simply a slavish copy of an item in the St Paul's archive but made elsewhere for the cathedral's use, then more such legal volumes may have existed elsewhere in early England—some perhaps written before the Conquest—and these simply do not survive.

§21.  This brief study of the annotations in CCCC 383 made by the Scipmen scribe and others is necessarily more suggestive than conclusive, and much work remains to be done on this manuscript. I hope, nonetheless, to have illustrated two points. First, Corpus 383 is an important manuscript for our understanding of the use and transmission of legal texts in early England and, although the manuscript generally and the Scipmen scribe's additions in particular have received quite a bit of scholarly attention, both could be better understood. In particular, the annotations in this manuscript are as important as the main text for understanding when and where the manuscript may have been compiled and how it was used. This leads to my second point—that even minor annotations and corrections in manuscripts generally and legal manuscripts in particular are an important source of information about those texts and should not be ignored. Annotations are physical evidence of readership and, where they are datable or localizable, can provide an historical context for the transmission and reception of texts. At a time when the corpus of Anglo-Saxon laws is attracting renewed interest and when new editions of many legal texts are being planned, I would urge editors not to ignore minor annotations and corrections, even where their content does not appear to affect the meaning of the text. In the case of CCCC 383 and other manuscripts, the mere physical presence of annotations can be as meaningful as their content.25


1.   Richards dates the manuscript to the late eleventh century and the Scipmen scribe's additions to the first quarter of the twelfth century; Keynes dates the manuscript as 'c. 1100'.  [Back]

2.   Note that on p. 165, however, Wormald seems to accept Ker's date of s. xi/xii; it may be that, on p. 228, he is simply stretching the boundaries of this date in order to emphasise its contemporaneousness with Textus Roffensis and Quadripatitus.  [Back]

3.   Recently, Thomas Gobbitt has suggested that both the production and annotation of the manuscript might have taken place in the context of the episcopacies of Bishop Maurice (c. 1085–1107) or Bishop Richard de Beaumis (1108–27) (Gobbitt 2009). I would lean toward the earlier of these historical contexts for the manuscript's production, but have no argument with either context.  [Back]

4.   So, for example, when Wormald notes that 'there are wildly inconsistent spellings, renditions of "king" ranging from "kyning" through to "cing", and not infrequently within a few lines of one another', he is in fact citing one of the most common features of eleventh-century English orthography. There are over forty distinct spellings of 'king' attested in eleventh-century English manuscripts, often alongside one another, so that even Cambridge, Corpus Christi College 162—an early eleventh-century manuscript written by a single careful scribe whose command of English could hardly be questioned—contains the syncopated form 'cyng' alongside the more dominant 'cyning'. For statistics on attested spellings of 'king' and on late Old English spellings generally, see Scragg et al. 2005.  [Back]

5.   See, for example, p. 32, line 15, where the main scribe writes 'ðeðearf' for 'beðearf', or p. 46, line 4, 'unfæð ða' for 'unfægða'.  [Back]

6.   BL, Royal 7. C. XII is the manuscript of Ælfric's homilies upon which Peter Clemoes based his edition of the first series of Catholic Homilies (Clemoes 1997). It is written primarily by two scribes who are thought to be working at Cerne Abbas under Ælfric's direction, and there are corrections in the margins which are thought to be written in Ælfric's own hand (Eliason and Clemoes 1966, 19–22, esp. p. 19, n. 8).  [Back]

7.   The presence of two quite different copies of A Gu, together with the pattern of corrections made by both the rubricator and the Scipmen scribe, lead me to surmise that CCCC 383 was compiled from more than one exemplar. On the corrections and their implications, see §15 ff.  [Back]

8.   See Parker on the Web website (  [Back]

9.   See Parker on the Web.  [Back]

10.   Wormald takes these red brackets to be clause divisions (Wormald 1999, 235). It is worth noting, however, that they do not always occur at the plausible beginning of a clause, and sometimes even occur in the middle of a sentence. I think it more likely that they are drawing the reader's attention to the subject of a clause, rather than marking the precise location of clause divisions.  [Back]

11.   For editions of the text, see Robertson 1939, no. 72, pp. 144–5; Kelly 2004, no. 25, pp. 192–201. Pamela Taylor usefully provides modern equivalents of the place-names in Taylor 2004, 15. For commentary, see Robertson 1939, 389–92; Taylor 1992; Taylor 2004, 15.  [Back]

12.   On the subject of naval taxes, see Lawson 1984.  [Back]

13.   Some of the place-names are listed in pairs, followed by a single number (e.g., 'Of Dunmæwan 7 of Tollesfuntan .i.'). As Pamela Taylor suggests, if one interprets these pairings to mean that each place must provision that number of men (rather than that they are together responsible for that number of men), then the total number of men to be provisioned by St Paul's according to this text is 58, and the discrepancy can be understood in terms of a scribal omission of a single minim (Taylor 1992, 298). On the possible sizes of longships of Æthelred's reign see Keynes 1980, 225, esp. n. 257.  [Back]

14.   The Anglo-Saxon Chronicle notes the destruction of St Paul's by fire in 962 (ASC MS A, an. 962) and again in 1087 (ASC MS E, an 1087). During rebuilding, much of the church was again lost to fire in the 1130s. Subsequently, the spire was struck by lightning in 1444, though not completely destroyed by fire until 1561, and the whole of the cathedral was again lost in the Great Fire of 1666 (Cragoe 2004).  [Back]

15.   See Parker on the Web.  [Back]

16.   See Parker on the Web.  [Back]

17.   Here, Liebermann incorrectly identifies the correcting hand as '16. Jh.', i.e., a sixteenth-century hand. He seems to misidentify and conflate several of the medieval annotating hands in this way, including both the rubricator's and the Scipmen scribe's. For other examples of the rubricator correcting instances of eyeskip in the main scribe's work, see CCCC 383, p. 1, line 20 (Hu 0–1); p. 41, line 4 (Ine 70:1); p. 45, line 24 (I Cn 22:1); p. 54, line 19 (II Cn 24:3); p. 56, line 11 (II Cn 30:1).  [Back]

18.   One could read the clause to mean something like, 'If an Englishman brings forth the charge [of cattle theft], he [the accused] will then deny the charge by [an oath of] double value; if it [the charge] is not then brought forth, the oath [is] not increased'. Obviously, the clause becomes much more meaningful with the addition of the missing words, but it still makes a kind of sense without them, even if it seems merely to be stating the obvious.  [Back]

19.   On Textus Roffensis and its relationship to CCCC 383, see Liebermann 1903–16, I.xxvi–xxviii; Liebermann 1898; Wormald 1999, 244–53. The manuscript is edited in facsimile in Sawyer 1957 and 1962. It is described in Ker 1957, no. 373.  [Back]

20.   'And likewise regarding the king's treasurers or any of our reeves who have been found guilty of stealing, [the penalty] will be the same for them', as opposed to the reading in TR, 'And likewise regarding the king's treasurers or similarly our reeves who have been found guilty . . .' See CCCC 383, p. 12, line 15 and Liebermann 1903–16, I.152 and n. 22.  [Back]

21.   'If he [a thief] should escape and hide, then he [the person who caught the thief] is responsible for the fine' (Liebermann 1903–16, I.100).  [Back]

22.   'If a man takes the charge for pasturage in pigs, [he should take] a third [of the pigs] at three fingers, a fourth at two fingers, a fifth at thumbs'. The Scipmen scribe's addition of the word 'spic' makes the various measurements in terms of fingers about the thickness of the bacon [spic] provided by the pigs.  [Back]

23.   Liebermann prints both versions in his Gesetze, referring to the official version (CCCC 383, p. 83) as B and the London version (CCCC 383, p. 6) as B2 (Liebermann 1903–16, I.126). Simon Keynes points out the London perspective of B2 (Keynes 1990, 233–4).  [Back]

24.   The view is typified by Patrick Wormald; see, e.g., Wormald 1999, 477–83, esp. 482–3.  [Back]

25.   My thanks are due to the Master and Fellows of Corpus Christi College for allowing me access to the manuscript, as well as to Ms Gill Cannell, Dr Suzanne Paul, and the rest of the staff at the Parker Library for being so accommodating while I was working there. I would also like to thank Dr Elizabeth Boyle and Dr David Woodman for the opportunity to present this research in a Research Seminar held in the University of Cambridge's Department of Anglo-Saxon, Norse and Celtic in June 2010, as well as all those who attended and provided useful feedback and comments. Any faults that remain are my own.  [Back]

Works Cited

Bately, J. M., ed. 1986. The Anglo-Saxon chronicle: a collaborative edition, MS A. Cambridge: D. S. Brewer.  [Back]

Clemoes, P., ed. 1997. Ælfric's Catholic homilies: the first series. EETS s.s. 17. Oxford: Oxford UP.  [Back]

Cragoe, C. D. 2004. Fabric, tombs and precinct 1087–1540. In St Paul's: the cathedral church of London, 604–2004, eds. D. Keene, A. Burns and A. Saint. New Haven and London: Yale UP.  [Back]

Eliason, N., and P. Clemoes, eds. 1966. Ælfric's first series of Catholic homilies: British Museum Royal 7 C. XII, Fols. 4–218. EEMF XIII. Copenhagen: Rosenkilde and Bagger.  [Back]

Gobbitt, T., 2009. Audience and amendment of Cambridge, Corpus Christi College 383 in the first half of the twelfth century. Skepsi 2:6–22.  [Back]

Irvine, S., ed. 2004. The Anglo-Saxon chronicle: a collaborative edition, MS E. Cambridge: D. S. Brewer.  [Back]

Kelly, S., ed. 2004. Charters of St Paul's, London. Anglo-Saxon Charters 10. Oxford: Oxford UP.  [Back]

Ker, N. R. 1957. Catalogue of manuscripts containing Anglo-Saxon. Oxford: Clarendon.  [Back]

Keynes, S. 1980. The diplomas of King Æthelred the Unready, 978–1016: a study in their use as historical evidence. Cambridge: Cambridge UP.  [Back]

———. 1990. Royal government and the written word in late Anglo-Saxon England. In The uses of literacy in early medieval Europe, ed. R. McKitterick. Cambridge: Cambridge UP.  [Back]

Lawson, M. K. 1984. The collection of Dangeld and Heregeld in the reigns of Æthelred II and Cnut. English Historical Review 99:721–38.  [Back]

Liebermann, F. (with A. A. Arnold). 1898. Notes on the Textus Roffensis. Archaeologia Cantiana 23:94–112.  [Back]

Liebermann, F. 1900. Matrosenstellung aus Landguetern der Kirche London um 1000. Archiv für das Studium der Neuren Sprachen und Literaturen 104:17–24.  [Back]

———. 1903–16. Die Gesetze der Angelsachsischen. 3 vols. Halle: M. Niemeyer.  [Back]

Richards, M. P. 1986. The manuscript contexts of the Old English laws: tradition and innovation. In Studies in earlier Old English prose, ed. P. E. Szarmach. Albany: SUNY Press.  [Back]

Robertson, A. J., ed. 1939. Anglo-Saxon charters. Cambridge: Cambridge UP.  [Back]

Sawyer, P. H., ed. 1957 and 1962. Textus Roffensis: Rochester Cathedral Library Manuscript A. 3. 5. 2 vols. EEMF VII, XI. Copenhagen: Rosenkilde and Bagger.  [Back]

Scragg, D. G., et al., eds. 2005. MANCASS C11 Database [web database]. Manchester Centre for Anglo-Saxon Studies, University of Manchester.  [Back]

Swan, M., et al., eds. 2010. The production and use of English manuscripts, 1060–1220 [website].  [Back]

Taylor, P. 1992. The endowment and military obligations of the See of London: a reassessment of three sources. Anglo-Norman Studies 14:287–312.  [Back]

———. 2004. Foundation and endowment: St Paul's and the English kingdoms, 604–1087. In St Paul's: the cathedral church of London, 604–2004, eds. D. Keene, A. Burns and A. Saint. New Haven and London: Yale UP.  [Back]

Wormald, P., 1999. The making of English law: King Alfred to the twelfth century. Vol. I, Legislation and its Limits. Oxford: Blackwell.  [Back]