When the collaborative edition of the Anglo-Saxon Chronicle was inaugurated in the early 1980s with the publication of volume 4 (Simon Taylor's edition of MS B) in 1983, followed by volume 3 (my own edition of MS A) in 1986, the intention of David Dumville and Simon Keynes, the general editors of this highly ambitious and imaginative project, was clearly and carefully set out:
It is essential at the outset to provide reliable texts of / the vernacular annals in MSS. A-G, and (as far as / possible) of the several Latin works, in editions which / treat the individual manuscript or work in its own right; / each of these editions will be published in a separate / volume..., to facilitate comparison and so to aid the / process of establishing both how the manuscripts or works / differ and what they share. One can then proceed to the / production of editions of the constituent elements of the / Chronicle as a whole. These will comprise first and / foremost (as volumes 1-2) the main continuous chronicles: / that is, the "common stock" itself...to 892, and the so- / called "Northern recension".... The shorter identifiable / elements of the composite Chronicle will also be treated / separately (in volumes 10-14): these comprise the various / sets of annals which cover particular sections of the / period from the late ninth to the mid-twelfth century, / whether conceived as continuations of an existing series / of annals or as self-contained chronicles which came to be / associated with the rest. None of these... survives in its / original form, so that each has to be reconstructed from / the available witnesses and then judged on its own terms / (vol 3, vii).
One of these reconstructions, with the title The Abingdon Chronicle. A.D. 956-1066 (MS. C, with ref. to BDE) was scheduled to appear as volume 12, following sequentially The Chronicle of AEthelred and Cnut, A.D. 983-1022 (MSS. CDE[F]), the projected volume 11. The intention, then, was to begin with the publication of the texts of all seven manuscripts A-G in semi-diplomatic editions, with strictly limited coverage of content and relationships, and subsequently to use these editions as a starting-point for reconstructions of perceived constituent elements, accompanied by detailed discussions, speculations and arguments. In the event, however, only two of the semi- diplomatic editions have so far appeared,1and volume 5, dealing with MS C, the crucial text for the Abingdon Chronicle, is, thanks to a change of editor, not yet ready for publication. As a result Patrick Conner has been faced not only with a brief already precisely defined by the general editors, but with the very awkward task of having to anticipate and print in his volume a great deal of material which belongs properly to the edition of MS C and which will necessarily have to be covered a second time there, whether with the same or with different conclusions. He has also had to cope with the fact that a large part of the annal-material in MSS C, D and E for the prescribed period, 956-1066, has been earmarked for separate treatment and so placed out of bounds to him. Publication of the proposed reconstruction of this "Chronicle of AEthelred and Cnut" was, as we have seen, originally intended to precede Conner's volume. Its delayed appearance means that he is prevented from evaluating the theories that will be put forward there, not least those (presumably) arguing against an Abingdon origin for this very considerable section of the "Abingdon Chronicle".
At the same time Professor Conner has other problems unrelated to the series' chronology but raised by his brief. "An editorial reconstruction" he writes in his preface, "is essentially an unnatural textual act, and readers who misunderstand that put themselves in danger of being seriously misled by this edition. The Abingdon Chronicle apparently no longer survives, and its base text has to be inferred from the more complex...MS C.... It is even possible that there was never a manuscript which contained what I represent here as Abingdon's Chronicle, but only a sequence of texts which drew in different ways upon material produced at Abingdon at different times." And he goes on to explain that the text that he presents in his volume is "a model in the sense scientists use that word: it is the result of the application of a theory for the purpose of testing it". His text, therefore "has no authority distinct and separate from the discussion in the 'Introduction'". His expressed hope is that his comments there "will justify our reading a part of the Anglo-Saxon Chronicle profitably from the perspective of Abingdon Abbey."
The earliest surviving claim for an Abingdon origin for any part of the Chronicle was made by Archbishop Parker's secretary Joscelyn in the sixteenth century, when he described MS C as "Chronica Saxonica Abbindoniae", and Joscelyn's attribution has generally found acceptance.2There is no means of telling whether Joscelyn gave the MS this title on the basis of some external evidence now lost to us, or whether (and more probably) he was indulging in surmise, prompted by the same internal references to Abingdon affairs that have led modern editors and critics to suppose an Abingdon connection. If the latter, then, as Conner observes, the title is not without its difficulties, not least in respect of the expectations to which it is likely to give rise:
The student who comes to this edition largely unacquainted / with the textual problems of the Anglo-Saxon Chronicle / might expect to find a record of the local history of the / monastery at Abingdon and its dependencies and / temporalities. In fact, the material edited here is / concerned primarily with events and persons of national / prominence, and given the way MS. C seems to have / influenced other Chronicle manuscripts, two of which-- / MSS. DE--were never at Abingdon, Abingdon's recension of / the Anglo-Saxon Chronicle would seem to represent a / national chronicle.... Nor can one assume that everything / edited here as part of the Abingdon Chronicle entered the / complex tradition of the Anglo-Saxon Chronicle at / Abingdon, for in several instances it is impossible to / know to what degree a monk at Abingdon composed or / compiled annals afresh, or redacted or copied them from / other annalistic materials obtained from elsewhere (p. xi).
In his edition Conner begins with a brief examination of what is known today about the Anglo-Saxon abbey at Abingdon, his bibliography for this subject including an as yet unpublished article by David Dumville on the scriptorium and library of the abbey.3One historical event in particular attracts his attention: the restoration of the abbey in 956 after a period of apparent desertion and dereliction following the death of King Edmund in 946. It makes good sense for Conner, treading in the footsteps of scholars such as Dorothy Whitelock, to accept 956 as the most probable terminus ante quem non for the composition of the hypothetical Abingdon chronicle as here defined (we have to wait for the publication of the reconstruction of the "common stock" of the Anglo-Saxon Chronicle to be told the reasons why, pace Plummer, the "pre- reformation" Abingdon Abbey is not a candidate for the origins of this component also). However, I am unhappy with what I take to be the underlying implications of Conner's cautious comment that "[it] is noteworthy that, during this period, neither of the Anglo-Saxon Chronicle manuscripts identified with Abingdon--that is, MSS. B and C--contains any commentary for the period between 946 and 955" (p. xv).
Annal 946 is certainly the last of a group of annals with definable characteristics, and had the argument been that this group was composed at Abingdon, then it might conceivably have been argued that the absence of material for the years 947-955 was the direct result of the "apparent desertion" of the abbey during that period. However, this is not Conner's argument (and I think rightly so). Moreover, although MSS D and E have some material referring to northern affairs in annals for 947, 948, 949, 952 an d 954, MS A at this point merely records one event--the death of the bishop of Winchester--in an annal (dated 951) which is written in a hand which differs in several respects from that of the main scribe of this section and which indeed may well be a later addition.4Any argument ex silencio must surely be accompanied by details of the material that we might in other circumstances have expected an Abingdon chronicler to include, and perhaps even examine the reasons why MS. A likewise falls essentially silent at this same point.5And were contrary arguments being sought, it might be observed also that it is no less noteworthy that no reference is made to the appointment of the great scholar AEthelwold as abbot of Abingdon in 956 (or indeed to his departure in 963). However, as we have seen, Conner's brief is to argue the case for an Abingdon Chronicle from 956 and, in testing the theory, to separate what extant material might be said to support it from that which does not.
It is this material which is the subject of the section of the book that follows. This begins with brief introductions to MSS B, D and E--three of the four Chronicle manuscripts containing texts derived from the hypothetical single source- -followed by a very full examination of the codicology of the MS. C codex and finally a short account of two twelfth- century cartularies from Abingdon which contain material relating to events of the late tenth century.
The account of MS. C is almost fourteen pages long and since it anticipates Katherine O'Brien O'Keeffe's forthcoming edition of C and any new or potentially controversial points will have to be fully assessed there, I shall pause here only to comment on what seems to me a most helpful innovation in the presentation of Chronicle material, a table of annal- groupings, or "segments", which Conner (p. xxi) sees as "much like the 'fragments' or repeated groupings of tales in the manuscripts of The Canterbury Tales". I found this table invaluable in following through his subsequent discussions and analyses of the available materials, though some of the divisions appear more arbitrary and open to debate than others. So, for instance, Segment 2 runs from annal 956 to 982 at 143v, where Connor sees a change of scribe in the middle of a sentence. The accompanying notes appear to imply that all these annals are in MS B as well as in MS C, though elsewhere in the edition the fact that MS B ends with annal 977 is fully covered and indeed on p. xxxix we are warned that the segmentation adopted oversimplifies the evidence because it "suggests that a hard line can be drawn between segments 2 and 3". As for the change of scribe, I find Conner's arguments convincing (just as I accept his explanation of an apparent break in MS C at the point where MS B ends);6indeed not only are the differences of hand not confined to the ones that he lists, but there are also changes at this point in the representation and position of the annal numbers.7However, an earlier indisputable change of scribe is not reflected by a corresponding change in segment-divisions, while the annals in this segment that are "rhetorically unlike earlier work" and which as a group "focus on AEthelred's Danish campaigns" do not begin till 991. Also breaking new ground is the suggestion that the twelfth-century scribe who completed the annal for 1066 was not adding to a manuscript from which one or more leaves had been lost (as earlier scholars such as Dorothy Whitelock8and Audrey Meaney had argued) but substituting a cancel for a worn-out leaf and copying onto it the material originally entered by Conner's scribe 8. The section on materials, with its examination of the whole of the C-text in segments, is followed by a detailed, thorough and careful examination of the "text-history of the Abingdon Chronicle". Here the segmental divisions are replaced by divisions according to annal date and instead of beginning with 60 B.C. start with the annals for 956 to 1018 (corresponding to MS C segment 2 and part of segment 3). This section begins with an important statement, "justified", in Conner's words, by the "physical evidence of MS. C and the comparative evidence provided by MSS. BDE":
In or about 1044, much of the extant MS. C was written / for Abingdon, probably as a project for the house; the / copyists probably compiled MS. C from several sources / which did not include MS. B (p. xxxiv).
So in one sense at least MS C is the Abingdon Chronicle. As Conner observes in his summaries, p. lxxi: "The Chronicle as we have it in C...must have been written in late 1043 x 1044". (Entries that follow after this date in MS C are taken to be Abingdon continuations, with annals from 1044 to 1048, for instance, added separately and being "presumably contemporary with the events they describe".9However, they are printed in the reconstructed text as an integral part of the Abingdon Chronicle, presumably because this formed part of the editor's brief.) At the same time Conner recognises that MS C's sources included a hypothetical BC and CE, the former "a working copy which", he believes, "continued to be updated as much as five years after B was copied from it, and which was copied again to produce C", the latter, with annals from 983 to 1048, the "probable copy-text for C through annal 1042 or perhaps 1043" (p. lxx). Both of these hypothetical MSS are assigned an Abingdon origin. An unnumbered table (pp. lxxii- lxxxi) indicates the evidence on which each of seventy-one annals in C may be considered an Abingdon product.
After the "Introduction" come an excellent "Bibliography", the "Reconstructed Text" and finally, following the pattern established for the series by Simon Taylor and subsequently slightly modified by me, "Indices of Personal Names and Places". (Incidentally, I dislike the heading "Index of personal names: Insular" as much now, as I did at the time of publication of my edition of MS A, when I had to include under this heading not only William the Conqueror (as here) but also Julius Caesar and the Greek Archbishop of Canterbury, Theodore.) My quibbles about Conner's text relate essentially to matters beyond the individual editor's control, as for instance the emasculation necessarily resulting from the prescribed exclusion of material to be published as part of a "Chronicle of AEthelred and Cnut",11and, in the footnotes, the practice of separating MS variant from MS name by a comma, as, e.g. [1043] 5: MS. E, Winceastre, beside [1043] 3: MS.D, cynge; E cyng, and the extent of coverage there of details specific to MS C, with, for instance, comments such as p. 3, annal 971 note 2: "The scribe used a point here, possibly to set off Oskitel arcebisceop". Certainly, given that MS C has been taken as base text, any emendations to C should be backed up by textual notes. But unless MS C is assumed actually to be the Abingdon Chronicle then manuscript peculiarities of this type are surely not relevant. And incidentally, unless C is supposed to be the source of the texts in D and E (and emendation of MS. C annal 1040 indicates that on occasion they have the expected reading where MS. C is defective), then some consideration ought to have been given to the possibility that at least a few of the variants in those MSS other than the entry s.a. 1018, might represent the reading of the lost hypothetical original. In the case of MSS D and E variants are cited in the footnotes; in the case of MS B they are not recorded at all.
The arguments that Conner produces are too many, too dense and too complex to be explored in depth in a review and necessarily involve a great deal of speculation: in any case they will need to be given very serious attention by the editor of MS C, whose task I would not wish to complicate further by indulging in my own dialogue. However, I should like to make a few comments on the following details:
1. p. xxxvii:
As I have pointed out elsewhere,11on is a not uncommon variant of ond in Old English manuscripts (see now also the Toronto Dictionary of Old English entry for And). The case for C's having used an exemplar other than B at this point, however, does not depend on this "further support".
2. p. xxxviii:
While caution requires us to remember the possibility / that C may have shifted exemplars after 652 and again / after 946, the more probable turn of events for a / manuscript being written as late as 1044 is that it / adhered to one exemplar which, given the similarity of / texts, is surely the same exemplar B used. If that / exemplar, BC, included a cramped, undated / collection of texts for the period 653 to 946,12then / it would have been thought appropriate to annotate these / with annal numbers before copying them, and perhaps to / smooth them out and annotate them in other ways as well. / We can only conjecture at this point where a scribe got / the dates....
All material in C which is not found in B [in the / section 653-946] is easily accounted for by remembering / that sixty-six years passed between the copying of B / and C; the notion that BC was augmented and / annotated during that time is sufficient to explain / additions, such as annals 675 and 976, which B has not / got.
I appreciate Conner's note of caution here as well as the neatness of his proposed solution to the problems posed by this part of the Chronicle. However, my doubts about C's use of B's immediate exemplar for the section 653-94213remain. Clearly it must be left the editor of MS C to evaluate the strength of our respective theories and supporting arguments.
3. pp. xxxviii-xxxix:
The annal for 971 is the first text after 946 which is / complex enough to have been composed by a chronicler, / and it started on a fresh line without an annal number, / because the numeral was listed as the last of a series / of barren numerals for 960 to 971; the annals for 956, / 957, and 959 record only royal obits and subsequent / successions, and thus they may constitute "backfill", / added to BC at Abingdon as an attempt was made to / fill in the space between the earlier stock of BC / and a later collection of materials added to BC / preparatory to copying MS. B in 977. Actually, barren / numerals in this part of the chronicle begin with 947, / but entries come nine years later, concurrent with the / time at which AEthelwold established his reformed / monastery there....
I have already commented on the difficulties inherent in an argument ex silencio, with respect to the period 946-955. That the annals 956, 957 and 959 record only royal obits etc. is a fact that has to be interpreted with equal care, given that (i) as Conner has reminded us, this is a national not a local Chronicle and (ii) MS A at this point has entries only for 955 and 958, covering precisely the same events as those in C 956 and 959 (death of King Eadred and accession of Eadwig; death of Eadwig and accession of Eadgar).14The scribe of A at this point was in Dumville's view (and I concur) making these entries more or less contemporaneously with the events described in them. As for the accompanying argument concerning the lack of complexity of these three annals, I cannot accept what appears to be the underlying assumption, namely that brief entries cannot be/are not likely to be, the work of a chronicler.15
4. p. xxxix, n. 80:
An hypothesis which argues that the exemplar was a / chronicle which, in 956 x 977, had not been updated / since 946 would fit the circumstances here. / Presumably, such a chronicle would have been brought to / Abingdon, and brought up to date. This may have / happened as late as 977, when all the material which is / in MS. B had been added to BC. MS B, a fair copy / of BC by a competent scribe, was made, we may / speculate, for presentation purposes to someone who / took no interest in it or who did not have access to a / clerk sufficiently skilled to keep it up. That would / both explain why Abingdon kept up BC rather than / adding to the new copy made in 977, and it would / explain why MS. B was not kept up. While this is, I / emphasise, purely speculative, it is also an / exceedingly economical explanation.
Professor Connor has been invited to reconstruct a hypothetical chronicle; he is therefore fully entitled to indulge in speculation. This can only be assessed in terms of probability (and the most probable is not necessarily also the correct explanation) and/or countered by still more speculation, which in its turn may or may not have a basis in fact. I applaud the neatness and indeed the economy of his hypotheses here, and I find the argument for a 977 updating most attractive--though I would reserve final judgement at this stage in Chronicle-studies. Furthermore that MS B comes to an end with the annal 977 and events during the lifetime of King Edward, while the accompanying regnal list goes no further than the accession of that king, who, both MS A and MS C report, was murdered in 978 makes it not implausible that B should have been copied (I) immediately or (2) shortly after BC entered 977.16It is also plausible to hypothesise as Conner does, p. xxxix, that the fact that C records the murder of King Edward under 978 and subsequent events as late as 982 in this segment, "probably signifies Abingdon's maintenance of annalistic activities in the tenth century at least until 982". Although Conner not unreasonably includes it and several successive annals in his text, 983 is, of course, the first of the annals earmarked by the general editors for the reconstruction of the Chronicle of AEthelred and Cnut.
5. p. xl-xli:
A particularly welcome new contribution to the study of the Anglo-Saxon Chronicle is Conner's use of passages from Abingdon cartularies to illuminate Abingdon annals. Of the relationship between a Latin text referring to the death of Sidemann and BC's entry for 977, Conner states that it is obvious that one is a translation of the other, either directly or through an intermediary text. I agree with his preference for C (or BC) as lying behind the Latin as the simpler of the two hypotheses which he puts forward. However, I wonder whether we can wholly rule out a third possibility, that both depend on a now lost record of the event--a possibility which he himself seems to recognise when he goes on to refer to "the Latin text or its source". The second cartulary passage is also of considerable interest. However, I cannot agree with the conclusions that Conner draws from what he sees as significant correspondences between the Sidemann obit in the Chronicle, "Her waes thaet myccle gemot aet Kyrt lingtune ofer Eastron, 7 thaer forthferde Sidemann bisceop...", and part of a sentence in what he calls the AElfhere Memorandum in the Abingdon cartulary, Cotton Claudius C. ix, which reads: "Tha waes ofer Eastron micel gemot aet Athelwarabirig, 7 hit wes gesitolad thar tham hlafardingan tha thaeron weron.... "17In both cases, he observes, an expletive construction ("tha waes" or "Her waes") controls the first clause and an adverbial ("thar" or "thaer") relates the second clause to it. Both describe the two councils in the same way ("micel gemot" or "myccle gemot"), "which in itself would not be significant except that both collocate these with the same temporal reference to Easter ('ofer Eastron')". He goes on to comment that this prepositional phrase "is not randomly distributed throughout the Old English corpus, which reduces the chance that its occurrence twice in these associated passages can be attributed to coincidence" (p. xliii).
It may well of course be that there is indeed some connection between the passages under scrutiny. However, it should be noted that this is not demonstrable from the constructions that Conner cites as significant or potentially significant. The collocation micel gemot is not confined to chronicle contexts--as a reference in the West Saxon Gospels (Matt. 26.4 Hi haefdon mycel gemot) demonstrates,18while the holding of a major council in the Easter period is, as a historical fact, in itself in no way a matter for surprise. As for the shared use of ofer Eastron, it is true that in surviving texts this collocation is "concentrated in the Anglo-Saxon Chronicle" and that "the distribution by manuscript favours MS C by three to two over D...nearly two to one over A, and three to one over E". However, the significance of these figures can only be assessed in the context of opportunity. An extension of the search to include all recorded Old English collocations for "after Easter" opens up a somewhat different perspective on Chronicle usage. First of all, it reveals that MSS A and B never use any expression other than ofer Eastron--and all the instances of this are shared with MS C. The alternative aefter Eastron19occurs once in MSS C, D and E in the annal for 1016 (part of the hypothetical "Chronicle of AEthelred and Cnut") and twice more in MS E-- once in a post-Conquest entry (1116), and once in the collocation thaes aefter Eastron, replacing A,B, C and D 871 thaes ofer Eastron. The collocation uppon Eastron is found only in MS E 1095, while the related construction with onufan (a preposition used with midne winter in the Mercian Register entries in MSS B, C and D for 915), is not recorded in any of the Chronicle texts. On this showing the collocation ofer Eastron is the normal rendering of the phrase after Easter in Chronicle texts right into the eleventh century, and its occurrence in material relating to the tenth century in both the Abingdon Chronicle and Abingdon memorandum has no special significance. At the same time, there is sufficient other evidence for the inclusion of dated references to important meetings or assemblies in charters such as the one under scrutiny--on occasion even introduced by the collocation tha waes,20to show that we are not compelled by content of this one to assume a chronicle entry as the source. For me the great importance of the cartulary material cited by Conner here and subsequently, in both Latin and Old English, is the evidence it affords for an active Abingdon scriptorium and for the (hypothetical but highly probable) availability at Abingdon of documents useful to chroniclers, whether as part of a "house narrative" or as individual records. Presumably Professor Dumville's forthcoming article will shed further light on this scriptorium. In the interim, I would ask (most tentatively) whether the undeniable relationship between the cartulary entry concerning the sudden death of Sidemann and annal 977 (the last to be entered in MS. B) might not support a hypothesis of a kind of "backfill" extending to the end of that annal!
6. Not the least important of Conner's hypotheses relates to the Chronicle poems. Of those at annals 974 and 975, known today as "The Coronation of Edgar" and "The Death of Edgar" he observes (p.lxxiii) that their texts appear in MS A in the hand of an early eleventh-century scribe, and nothing suggests that they originated in the MS A context. Therefore, their "earliest expression may be attributed to *BC which was copied into MS B at Abingdon, probably in 977/8; the poem[s], then, [were] composed [973 and 975] x 978, very possibly at Abingdon".21And again at pp. lxxxii-lxxxiii he suggests that every poem which is located within the Anglo-Saxon Chronicle (i.e."The Battle of Brunanburh", annal 937, and "The Capture of the Five Boroughs", annal 942, as well as poems in MS C at 1036 and 1065), is "to be found in MS C within an Abingdon context", although "it is almost impossible that all of these poems could have been composed by the same person". And he continues, "It is not... difficult to imagine that 'The Battle of Brunanburh' was inserted where someone felt a text should commemorate an event which, even if *BC had been updated as late as 977, would still have existed in living memory. If the update to *BC were made in 956 (as early as we might expect at Abingdon), then we might be even more inclined to imagine that the poem could have been composed when Abingdon was developing its recension of the Chronicle..." Until the publication of Conner's forthcoming article, "Poetry and the Institution: The Case for Abingdon", it is not possible properly to assess his claims here. However, although I suppose that major stylistic differences could be the result of changing fashion,22it is surely not just "almost impossible" but totally impossible for a single author to have been responsible both for the poem on Brunanburh (copied into MS A in the mid 950s and commemorating an event of 937) and that on the death of Edward the Confessor, 1065. (Incidentally the terminus ad quem for both "The Battle of Brunanburh" and "The Capture of the Five Boroughs" has to be mid 950s, not, as Conner, p.lxxxi, 977/8).23I would in my turn engage in speculation and ask whether the entries for 973 and 975 might not be the work of an author who was attempting to produce a continuous verse narrative in annal form, following the precedent of the earlier verse annals by incorporating the Chronicle "Her" formula, but going beyond the mere act of commemoration of a victorious campaign or royal deaths and successions.
7. Last and not least: it is essential for users of this volume not to forget that Conner's brief, which he has tackled with great determination, thoroughness24and ingenuity, was to assume the existence of an Abingdon Chronicle 956-1066. (Reading between the lines of some of his conjectures, this reviewer at least has a lurking suspicion that had the author been given a free hand he might have set himself very different parameters). He has accordingly set out to find evidence in support. It is perfectly possible that editors of future volumes in this series, surveying the Chronicle MSS from other perspectives, will come up with counter-arguments for at least part of the material, and indeed that the general editors in their final over-view will seek to produce modifications and reassessments of their preliminary positions--the whole point after all of the series. In his "Introduction" Patrick Conner has very fully tested his initial theory. That a version of the chronicle was made and kept up at Abingdon is the most economical and most satisfactory explanation for the presence of entries containing references to Abingdon in some of the surviving Chronicle MSS. However, as he scrupulously reminds us, the exclusions of certain materials that he has been obliged to make may have resulted in the production of "a text which the chronicler himself did not make and would not have made" (p. xii). A number of his conjectures and hypotheses, however reasonable, are capable of being countered by no less reasonable alternatives and at this stage--perhaps any stage --in Chronicle studies, as with a complex jigsaw from which many pieces are missing, we cannot hope for certainty.