A facsimile edition of the Annals of Roscrea
Bart Jaski and Daniel Mc Carthy
Abstract
The Irish chronicle known to modern scholarship as the ‘Annals of Roscrea’ is found only in the manuscript Brussels,Bibliothèque Royale 5301-20 pp. 97−161. It was first registered in print in the comprehensive catalogue of the manuscripts in the Burgundian Library at Brussels published in 1842, and an edition was published by Dermot Gleeson and Seán Mac Airt in 1959. Recent research has shown that the principal scribe, the Franciscan friar Fr Brendan O’Conor, transcribed his source, ‘mutila Historia D. Cantwelij’, in two successive phases and then in a third phase it was annotated and indexed by his fellow Franciscan Fr Thomas O’Sheerin. This research has also shown that the edition of Gleeson and Mac Airt is incomplete, having omitted the pre-Patrician section of the chronicle. Hence this, the first full edition of the work, has been prepared in facsimile form so as to make clear the successive phases of compilation of the text, to provide an accurate account of its orthography, to identify the relationship of its entries to those of other chronicles, and to furnish an AD chronology consistent with the other Clonmacnoise group chronicles.
Introduction
The sixty-five pages of the composite manuscript,Brussels, Bibliothèque Royale 5301−20 pp. 97−161, contain a chronicle in Latin and Irish written by the Franciscan friar, Fr Brendan O’Conor. It is virtually certain that O’Conor transcribed this chronicle in London in July 1641 from an exemplar then in the possession of Finghín Mac Carthaigh, alias Florence Mac Carthy. Subsequently O’Conor’s transcription was known in Louvain to his Franciscan contemporaries, Fr John Colgan †1658 and Fr Thomas O’Sheerin †1673, and a substantial index to it was compiled by O’Sheerin.[1]However, after O’Sheerin’s work we have no further reference to this chronicle until 1842 when it was recorded in the comprehensive catalogue of the manuscripts in the Burgundian Library in Brussels that was published under the direction of J. Marchal, ‘le conservateur des manuscrits de l’État’.[2] Two years later, Laurence Waldron, at the instigation of Eugene O’Curry, re-discovered the Franciscan manuscripts in the Burgundian Library in Brussels. Two years later again Samuel Bindon compiled a short catalogue of the manuscripts of Irish interest which was first published in 1847 in the PRIA on the initiative of James H. Todd, and very shortly afterwards Bindon also published the catalogue independently. These two publications of Bindon’s catalogue of Burgundian manuscripts are virtually verbatim and in the latter he gratefully acknowledged the RIA’s permission ‘to get a few additional copies struck off’.[3]However, even after publication of its existence this chronicle went virtually unremarked for over a century until 1959 when Dermot Gleeson and Seán Mac Airt published an edition of the post-Patrician section in the PRIA.[4]Since then the ‘Annals of Roscrea’ (AR) have been regularly mentioned in most serious discussionsof Irish Annals; for example, Mac Niocall in 1975, Grabowski and Dumville in 1984, Mc Carthy in 1998 and 2008, Charles-Edwards in 2006, and Evans in 2010.[5]Most of these authors have recognised the close relationship between the content and organisation of AR and that of the Annals of Tigernach (AT) and Chronicum Scotorum (CS), and hence most have classifed AR as a member of the Clonmacnoise group of annals.[6] However, in some of these publications uncertainty has been expressed regarding both the extent of the chronicle ‘Annales Roscreensis’, and the exemplar from which it was drawn, and it is to this matter that we now turn.[7]
Scope and origin of the title ‘Annales Roscreenses’ and its exemplar
In a subsequent addition made on the upper margin of p. 1 of his transcription O’Conor briefly described his transcription and exemplar as ‘Adversaria rerum Hibernij quae excerpta ex mutila Historia D. Cantwelij’, that is, ‘Memoranda of Irish affairs excerpted from the mutilated History of D. Cantwel’. O’Conor also inscribed the title ‘Annales Roscreensis’ on his transcription, and, either this title, or its English translation, have been regularly used to designate this chronicle ever since. However, there has not been agreement amongst modern scholars either as to whether this title refers to the entire sixty-five page transcription, or whether the ‘Historia D. Cantwelij’ served as exemplar for the entire sixty-five pages. The reasons for this confusion and its resolution become quite clear when the manuscriptitself is examined for it then emerges that O’Conor, who himself paginated the entire sixty-five pages as pp. 1−65, first inscribed ‘Annales Roscreensis’ on the upper left-hand margin of p. 1, but then subsequently cancelled this and wrote it in very large letters on the upper margin of p. 25, immediately before the account of S. Patrick’s mission to Ireland. Thus the most prominent appearance of this title in the manuscript is that at the head of the post-Patrician section. Indeed O’Conor’s cancelled inscription on p. 1 has never been acknowledged in modern times. Consequently most modern scholars have taken the title ‘Annales Roscreensis’ to refer only to the post-Patrician section, and they have expressed ambivalence regarding their relationship with the ‘Historia D. Cantwelij’. However, O’Conor’s action in placing the description ‘Adversaria … ex mutila Historia D. Cantwelij’ at the very head of his transcription and then numbering his pages serially pp. 1–65 shows both that he considered it a textual unity and that he had drawn all of these ‘Adversaria’ from the ‘Historia D. Cantwelij’. Furthermore collation of AR’s pre- and post-Patrician sections with AT/CS repeatedly discloses cognate entries throughout, and this independently confirms the unity of AR’s entire chronicle.
Moreover this view was certainly shared by O’Conor’s contemporary, O’Sheerin, who was responsible for the first stage in the compilation of the composite volume, now Brussels, Bibl. Royale 5301−20. In this compilation O’Sheerin originally assembled thirty-seven items into a single volume and prefixed to this a page listing the ‘Series hîc contentorum’ in which these items were registered under twenty-three headingsenumerated ‘1’–’23’.The hand of this ‘Series hîc contentorum’ and the indices to FA and AR is established as that of O’Sheerin by collation with his four signed letters to Francis Harold, MS Killiney D.5 pp. 9, 15–16, 177–8, 237.[8] In his compilation of this composite manuscript O’Sheerin originally placed AR first and explicitly stated its title and page count in his prefixed list of contents as ‘1. Annales Roscreenses per pag. 65’.[9] Subsequently, O’Sheerin, when he compiled his index to these annals, made absolutely explicit his view that ‘Annales Rosreenses’ referred to the whole chronicle and that its ‘extracta’ had all been drawn by O’Conor from ‘Historia D. Cantwelij’. For O’Sheerin commenced by cancelling O’Conor’s own heading ‘Index Annalium Roscreensium’on p. 163, and then wrote his own heading immediately below as follows:[10]
Extracta per Patrem Fratrem Brendanum Conorum ex Annalibus Roscreensibus seu Codice R.D. Cantwel, hîc digesta ordine Alphabetico, praetermissis tamen iis quae praecesserunt Missionem S. Patricii, annotatis ad marginem annis quibus quaeque acciderunt, juxta Annales Dungallenses.
Here by placing ‘seu’ between his references to ‘Annales Roscreenses’ and ‘Codex R.D. Cantwel’ O’Sheerin showed he considered them synonyms; by stating that these ‘extracta’ had been ‘hic digesta ordine Alphabetico’ he affirmed his authorship of the index; by remarking that the extracts preceding the mission of S. Patrick had been omitted from his index he made it absolutely clear that he considered that O’Conor had taken all of these ‘extracta’ from the codex of ‘R.D. Cantwel’.
O’Sheerin made another reference to ‘Annales Roscreenses’ when compiling his catalogue of the manuscripts and books in John Colgan’s study in Louvain following the latter’s death in 1658, now designated as MS UCD Killiney A34 item 1.[11] It is evident that O’Sheerin compiled this catalogue shortly after Colgan’s death because the catalogue gives precise locations on tablesand in pressesand chests for virtually all of the listed items. That it was in Colgan’s study is indicated by Bonaventure O’Docherty’s heading to his catalogue of c.1673, ‘Catalogus Manuscriptorum tam Latinè quam Hibernicè, olim in Camera R.P. Colgani repertorum, quibus postea R.P. Sirinus usus fuit’.[12]Under the heading of ‘Post praedicta [sc. manuscripta Latina]manent sequentia in mensa in fasciculus distinctis’ O’Sheerin included the item ‘De Hiberniae etcra quaedam ex Annalibus Roscreensibus, et alia Regulae diversorum Ssorum Hiberniae’.[13] While this entry makes no reference to either the scope or exemplar of the chronicle it does demonstrate that the expression ‘Annales Roscreenses’ was in use as a title in Louvain in Colgan’s time, and hence that Colgan knew the chronicle. Indeed, since we have seen above that O’Sheerin used this title as a synonym for ‘Codex R.D. Cantwel’ it seems most likely that his ‘quaedam ex Annalibus Roscreensibus’ actually refers to O’Conor’s transcription itself; certainly there is no other entry in the catalogue that could be considered to reference AR. Taken together O’Sheerin’s references to ‘Annales Roscreenses’ clearly demonstrate that he considered the title to designate O’Conor’s transcriptions from both the pre- and post-Patrician sections of the codex of R.D. Cantwel.We know of no other subsequent reference to ‘Annales Roscreenses’ from the context of Louvain; it does not appear, for example, in Bonaventure O’Docherty’s catalogue compiled evidently following O’Sheerin’s death in 1673.[14]However, under the heading ‘Catalogus Librorum in Camera R.P. Sirini repertorum praeter illos de quibus in praecedenti catologo’ O’Docherty entered the item, ‘Analecta de Rebus Hiberniae’, and this description would indeed accord with O’Sheerin’s composite volume including the ‘Annales Roscreenses’, now Brussels, Bibl. Royals 5301−20.[15]
Indeed, we know of no other references to ‘Annales Roscreenses’ until 1842 when a comprehensive catalogue of the manuscripts in the Burgundian library at Brussels was compiled under the direction of J. Marchal. In the first volume the contents of O’Sheerin’s composite volume were numbered as the twenty items 5301–20, and in most instances for each item was cited the names of the authors, incipit, language and its date. Our chronicle and O’Sheerin’s index were catalogued as items 5303–4 as follows:[16]
No / Noms des Auteurs / Incipit / Langue / Date ouSiècle
5303 / Cantwel – Adversaria rerum Hibernia / – / Latine / XVII 1/3
5304 / Brendani Conori – Extracta ex annalibus Roscreensibus / Adamnani abbatis / Latine-irl. / XVII 1/3
Here clearly the Burgundian cataloguer considered the chronicle a single textual entity drawn in the seventeenth century from the work of Cantwel, while he mistakenly characterised the subsequent index as simply ‘Extracta’ by Brendan O’Conor. Five years later in 1847 Samuel Bindon published his short catalogue of the books of Irish interest in the Burgundian Library in which, although he acknowledged the existence of the Burgundian catalogue, he gave no bibliographic details other than the following vague footnote: ‘The “Inventaire” is the first volume of the printed catalogue. In it the MSS. are enumerated without reference to subject; the second volume, or “Repertoire,” is a “Catalogue Methodique.”’[17]
Examination of Bindon’s catalogue shows that, while he regularly supplied additional details regarding the Irish manuscripts, these details are fairly frequently either inaccurate or inconsistent with the Burgundian catalogue. For examples: having stated that the volume contained ‘Nos 5301 to 5320, inclusive’, Bindon only gave identifiable accounts of 5301–14 and 5317–18, thereby omitting to register 5315–16 and 5319–20; he wrote that ‘5314 is an extract from Marianus Scotus’ whereas the Burgundian catalogue lists 5314 as ‘Martini Crusi – Extr. De annal. Suevicis’; Bindon was inaccurate in his identification of the number of manuscript folios and/or his citations of titles or incipits, and in particular his account of this chronicle and its index reads as follows:[18]
No. 5303 consists of sixty-five pages; the first twenty-six are entitled “Adversaria Rerum Hiberniae excerpta ex mutila Historia D. Cantwelly,” and commences thus: “Hoc anno ante diluvium.” At page 25 commences “Annales Roscreenses.” The initial line is “Patricius Archiepus in Hiberniam venit atque Scotos baptizare inchoat, nono anno Theodos. minoris,” &c. These Annals, as well as the “Adversaria,” are in Latin and Irish, and very badly written.
No. 5304 is a very long alphabetical Index of the Annals of Roscrea, made by “Frater Brendanus Conorus,” accompanied by marginal references to the Annals of Donegal.
Here Bindon’s citation ‘Adversaria … D. Cantwelly’ is both incomplete and orthographically inaccurate, and it was said to entitle only pp. 1–26. On the other hand the title ‘Annales Roscreenses’ was applied only to pp. 25–65, and the first line of p. 25, ‘Patricius Archiepus …’ described as the ‘initial line’. Thus Bindon divided the chronicle into two sections and incongruently placed pp. 25–6 in both sections. At the same time,while correctly identifying item 5304 as an index, he mistakenly attributed this to O’Conor. In this way Bindon’s catalogueeffectively restricted O’Conor’s identification of ‘Historia D. Cantwelij’ as his exemplar to just the pre-Patrician section, and restricted the title ‘Annales Roscreenses’ to the post-Patrician section, and misrepresented the authorship of the index. Most of these mistakes were repeated by Van den Gheyn in 1907 when he published a much more detailed catalogue of the contents of Bibl. Royale 5301−20.[19] Citing the PRIA publication of Bindon’s catalogue for ‘une analyse de ce volume par Bindon’, Van den Gheyn represented O’Conor’s transcription and O’Sheerin’s index as three distinct items as follows:[20]
6. (F. 51−76)Adversaria rerum Hibernie excerpta ex mutila historia D. Cantwelli. En irlandais et en latin.
7. (F. 76−83)[Annales Roscreenses]. Latin et irlandais.
8. (F. 84−119v)Extracta per Patrem fratrem Brendanum Conorum ex annalibus Roscreensibus seu codice R.D. Cantwel, hic digesta ordine alphabetico.
Thus Van den Gheyn, like Bindon, represented the ‘Adversaria …’ and ‘Annales Roscreenses’ as separate textual entities that incongruently shared ‘F. 76’, and then in contradiction of this representation he cited O’Sheerin’s heading to his index that asserted them to be identical. In this way the second and third published accounts of O’Conor’s transcription of ‘Historia D. Cantwelij’ erroneously restricted this source effectively to the pre-Patrician section of the text.
This confusion introduced by Bindon regarding the extent of the ‘Annales Roscreenses’ and the exemplar used by O’Conor had a serious consequence in 1959 when Gleeson and Mac Airt compiled their published edition of the text. In their description of the manuscriptthey followed Bindon and Van den Gheyn in designating the pre-Patrician section as the ‘Adversaria ... historia D. Cantwelli’ and asserted that ‘At p. 25 there commences the text of “Annales Roscreenses”’.[21] Since subsequent scholarship has referred to this published edition the consequence has been that the pre-Patrician section of the text has been effectively abandoned. For example, Grabowski and Dumville stated that, ‘The text of the annals [of Roscrea] is divided into four fragmentary series: (i) A.D. 432−40, (ii) A.D. 550–602, (iii) A.D. 440–77, (iv) A.D. 620−995’.[22]
Regarding the origin of the title ‘Annales Roscreenses’, it is the case that the earliest recorded instance of ‘Annales’ followed by a personal or place name used to entitle an Irish chronicle is that of James Ussher in 1609 referring to the manuscript, now TCD 1282, as ‘Annales Ultonienses’. Ussher’s student James Ware followed suiteover 1625–48 entitling other Irish chronicle texts as ‘Annales Tigernachus’, ‘Annales Inisfallenses’, ‘Annales de Loghkea’, ‘Annales Buellienses’ and ‘Annales Connachtus’.[23] The medieval Irish convention was to suffix a personal name to the words ‘leabhar’ and/or ‘airis’.[24] Now, as Gleeson and Mac Airt observed of AR, it is not the case that ‘the collection had any particular association with Roscrea’, so that it appears most likely that the title ‘Annales Roscreenses’ was in fact O’Conor’s own invention inspired by the Latin entitling conventions employed contemporaneously by Ussher and Ware, together with a knowledge of the Cantwel family’s association with Roscrea.[25]However, Ussher and Ware employed the word ‘annales’ to entitle chronicles that were substantially annual in character, as the word itself intimates. This suggests that O’Conor’s ambivalence in first inscribing his title on p. 1, and then cancelling that and inscribing it on p. 25,arose because the pre-Patrician section of his transcription is extremely intermittent as his own marginalia testify. Thus it appears that in relocating ‘Annales Roscreenses’ to p. 25 O’Conor was bringing his own nomenclature into accordance with the practice of Ussher and Ware. A title for this chronicle based upon its exemplar ‘Historia D. Cantwelij’ would in many ways be more appropriate and helpful, but since either ‘Annales Roscreensis’ or ‘Annals of Roscrea’ have been in use since c.1641, and are attached to the first published edition of the chronicle, it seems more practicable to retain these titles. However, it must be clearly understood that these titles apply to the entire sixty-five pages.
In this edition we shall use the siglum ‘AR’ prefixed to O’Conor’s page numbers 1–65 concatenated with the line numbers of his text to reference the entries, as will be explained in further detail below. In their edition Gleeson and Mac Airt also included O’Conor’s page numbers, and so these page and line number references may be used to readily locate entries in their edition. However, whenever it is necessary to refer precisely to entries in the edition of Gleeson and Mac Airt we shall use ‘AR’ followed by ‘§’ and their paragraph number. Thus, for example, AR 25.16 and AR §8 both refer to the entry ‘Natiuitas sanctae Brigidae’ commencing on the sixteenth line of p. 25.
Description of the manuscript[26]
Brussels, Bibliothèque Royale 5301–20 pp. 97–162 is a seventeenth-century Franciscan manuscript, and the only witness to AR. The published descriptions by the Burgundian Library, Bindon, Van den Gheyn, and Gleeson and Mac Airt are all brief and contain a considerable number of inaccuracies, and since we shall see that AR is an important witness to the Clonmacnoise group it is necessary to give here a detailed account of the manuscript and its text.[27] The manuscript was written by Fr Brendan O’Conor, a Franciscan friar who was sent from Louvain to Ireland in 1641 to collect historical material. It comprises thirty-three leaves measuring c.20.5×13.5 cm, and bears two paginations; the first, by O’Conor running pp. 1–65, was used by Gleeson and Mac Airt in their partial edition, and will be used in this edition.[28] The second, a modern pagination running pp. 97–162, is needed for references to the other texts in Bibl. Royale 5301–20. This volume consists of a compilation of over thirty Franciscan manuscripts, of which the first on pp. 1–70 is the only surviving copy of Dubhaltach Mac Fhirbhisigh’s Fragmentary Annals(FA),transcribed by O’Sheerin, Colgan’s successor in scholarship in Louvain, followed on pp. 71–88 by O’Sheerin’s alphabetic index to these annals.[29] Next, a letter by O’Conor on pp. 89–96 is followed by the text of AR on pp. 97–162, and its range is c.Flood–AD c.995, with lacunae at c.948 BC–AD 157, AD 252–335,480–549 and 602–619. The annals for AD 336–358 and 441–479 are also displaced, probably as a result of the mutilated state of its exemplar.The text of AR, like that of FA, is followed on pp. 163–234 by an index compiled by O’Sheerin, in which the personal and place names cited in AR following S. Patrick’s uenitare arranged alphabetically and indexed by O’Conor’s page number, and the chronology of events involving them is regularly tabulated ‘juxta Annales Dungallenses’. Thus the whole context of AR’s manuscript suggests an environment of intensive Annalistic study in Louvain, stimulated, no doubt, by the presence there of Michéal Ó Cléirigh’s compilation of the ‘Annales Dungallenses’, alias Annals of the Four Masters. In his heading to the index for AR O’Sheerin identified ‘Patrem Fratrem Brendanum Conorum’ as the scribe of extracts, ‘ex Annalibus Roscreensis seu Codice R.D. Cantwel’, and this identification is confirmed by comparing the Latin handwriting of AR with O’Conor’s letter which immediately precedes it. There is no date on the text of AR, but as it happens we do know something of the activities of Brendan O’Conor over 1641–2.