Entry point to the Scriptorium Bede knew at Wearmouth and Jarrow: The Canon Tables of the Codex Amiatinus
p. 187-197
Résumé
Ma communication entend mettre à profit le témoignage jusqu’ici inexploité des Tables des Canon du codex Amiatinus (Florence, Biblioteca Medicea Laurenziana, MS Amiatino, 1), en les utilisant comme une porte d’entrée dans le scriptorium du monastère double de Jarrow et Wearmouth. rédigé avant 716, le codex Amiatinus est comme une pierre angulaire dans l’étude des manuscrits du haut Moyen Âge, car il est localisé et daté avec précision. C’est le seul manuscrit subsistant des trois Bibles complètes commandées par l’abbé Céolfrith et à ce titre, c’est la plus ancienne copie complète subsistante de la Bible latine dans la version Vulgate. Dans leurs études consacrées au codex Amiatinus, R.L. S. Bruce-Mitford, Paul Meyvaert, Malcomb Parkes, Armando Petrucci et David Wright ont commenté astucieusement certains aspects du scriptorium northumbien où le manuscrit fut rédigé. Toutefois, les sept folios des Tables des Canons élégamment dessinées et enluminées et réellement utilisables (fol. 798r-801r) peuvent nous servir comme une autre porte d’entrée dans le Scriptorium de Jarrow et Wearmouth d’il y a presque treize siècles. D’après leur cadre architectural, leurs éléments décoratifs, le nombre de pages utilisées pour les tables, la distribution de celles-ci et le nombre de chiffres inclus dans chacune ainsi que l’ordre effectif de succession des chiffres et la comparaison de ces données avec celles des autres manuscrits contemporains, elles apportent un témoignage supplémentaire de la transmission complexe des modèles méditérranéens au nord de l’Angleterre et des préférences du scriptorium de Bède.
Texte intégral
Prior Scholarship on the Scriptorium of Wearmouth and Jarrow
1The Northumbrian scriptorium of Wearmouth and Jarrow maintains a respected, indeed a revered position, for scholars of many diverse disciplines. Bede himself comments in his Lives of the Abbots of Wearmouth and Jarrow on the importance placed on establishing a library by the double monastery’s founder, Benedict Biscop, who brought manuscripts from his many journeys to Rome, and subsequently established a scriptorium. Bede also relates that Ceolfrith, who became abbot of the double monasteries in 688, commissioned three complete Bibles, or pandects, a most unusual book-type for the late seventh and early eighth-centuries. We have confirmation that the scriptorium realized the completion of this phenomenal commission by 716. It was in this year, Bede tells us, that Ceolfrith resigned as abbot and left on his final pilgrimage to Rome, bequeathing two of the pandects to Wearmouth and Jarrow while carrying the only pandect to survive intact, now known to us as the Codex Amiatinus. With some variations this information is corroborated for us in another eighth-century text, Life of Ceolfrith, by an anonymous author. Although we have only a few tantalizing but important comments from Bede and the anonymous author on the pandects produced at the scriptorium, we also have surviving physical evidence. There is the one complete pandect, the Codex Amiatinus (Biblioteca Medicea Laurenziana, MS Amiatino 1) and several additional folia from one or both of the other pandects presently housed in the British Library1.
2Scholars have written directly or indirectly on various aspects of the scriptorium at Wearmouth and Jarrow providing an ever-expanding appreciation and awe of the tremendous energy, effort, expense, expertise, organization and training invested in the creation of a scriptorium that could fulfill Ceolfrith’s commission. Rosemary Cramp’s literally groundbreaking archaeological study of the monasteries in the 1960s and 1970s presented physical evidence of their structure, decorative elements and working implements, including writing implements2. Other scholars have focused their study on the surviving manuscripts to gain perspective on the organizing and the functioning of the scriptorium at Wearmouth and Jarrow. David Wright in his 1961 article in Traditio identified the hands of seven scribes who completed the nine sections of the Amiatinus3. Armando Petrucci’s discussion of the derivation of the uncial script used in the Amiatinus and M. Parkes’s 1982 Jarrow lecture, The Scriptorium of Wearmouth and Jarrow highlighted the achievements of the scriptorium and emphazised the astonishingly brief period of time from its founding to the production of manuscripts such as the Codex Amiatinus; Parkes especially pointed out the phenomenal quality of the early scribal efforts and the surprising degree of uniformity of script among the scribes indicative of the consistency of their training in comparison to the later adaptations at the scriptorium when the goal was to produce with speed the much demanded copies of the works of its ‘best- selling author’, Bede – particularly after his death. Richard Gameson’s fascinating study in Notes and Queries projected the underlying costs for the production of the three pandects giving us a better sense of the wealth of the double monasteries of Wearmouth and Jarrow and their level of commitment to providing the resources necessary for the completion of three complete Bibles. In addition, many scholars have addresssed aspects of the scriptorium while exploring the possible transmission of sources and the availability of models to the scriptorium at Wearmouth and Jarrow. In particular there is the much-debated issue as to the presence of one or more of the manuscripts from the sixth-century scriptorium, the Vivarium, established by Cassiodorus (c.480/5 - c.575/585) in southern Italy. Paul Meyvaert’s insightful 1996 article, «Bede, Cassiodorus and the Codex Amiatinus»4, is a recent contribution to this vexing question and the complexity of the transmission of sources.
3The light cast by these formidable and august scholarly investigations has benefited us all. My hope is that through the study of the canon tables within the Codex Amiatinus there might yet be further information to be gleaned regarding the scriptorium of Wearmouth and Jarrow. My presentation will present, describe and analyze the canon tables within the Codex Amiatinus resetting them within the codicological context of the manuscript, the Northumbrian context of the seventh and early eighth centuries and the broader Mediterranean context and will explore the possible source or sources of these tables.
Introduction to the Canon Tables of the Codex Amiatinus
4Within that treasure chest of a manuscript, the Codex Amiatinus, the canon tables dazzle and delight the viewer in an elegant, aesthetic tour de force on seven folia beginning at 798 recto and concluding on 801 recto. Yet, these tables are more than beautiful. They are a sophisticated, functional system of concordance used for cross-referencing textual passages between and among the four Gospels of the New Testatment as well as for identifying those passages with no correlations.
5Aspects of the canon tables of the Codex Amiatinus have been previously mentioned, but there is no study in which the canon tables are the primary focus. For example, R. L. S. Bruce-Mitford included a description of the art and the artists of these tables as part of his general discussion of the art in the Codex Amiatinus and compared them to the tables of the Lindisfarne Gospels (London, British Library, Cotton Nero D. IV) in the accompanying volume to the facsimile of that manuscript5. The canon tables were discussed in Carol Neumann de Vegvar’s appendix to her book, The Northumbrian Renaissance. Her focus was on the role the canon tables have played in the literature regarding which Cassiodorean manuscripts may have been available at Wearmouth and Jarrow6. Patrick McGurck also cited the canon tables of the Amiatinus in his broader discussion in «The disposition of numbers in latin Eusebian canon tables» where he cogently argued that noting the discrepancies and distribution of canon tables may be used for future diagnosis of family systems7. Even in Carl Nordenfalk’s seminal two volume work of 1938, Die spätantiken Kanontafeln, any mention of these canon tables is omitted because they are not within a Gospel book but within the rare book-type of a complete Bible8.
6Methodologically, study and discussion of these canon tables has been greatly limited by several factors – the fact that the Codex Amiatinus is a rare book-type and a singleton survivor in a field of study where surviving Gospel Books are numerous and maintain their stature in selection for study; the art and the script of the Codex Amiatinus are regarded as begin distinctly different from better known Insular manuscripts such as the Lindisfarne Gospels and the Book of Kells and treated as ‘other’; and for many years there has been the issue of access to the manuscript and the lack of good quality reproductions which has only recently been remedied by the publication of a facsimile and a CD rom.
7The canon tables of the Codex Amiatinus, this reservoir that has yet to be fully tapped, serve as our entry point into the scriptorium that Bede (d. 735) knew at Wearmouth and Jarrow. My study will recontextualize them on three different and distinct levels: first, the codicological context of the Codex Amiatinus; second, the Northumbrian context of the late seventh and early eighth centuries; and third, within the broader Mediterranean context and an exploration of the possible source of these tables.
Recontextualizing the Canon Tables of the Codex Amiatinus as an entry point to the Scriptorium Bede knew
8As we open the massive Codex Amiatinus we note that the text of the Old Testament ends with the Books of Maccabees I and II, and visually culminates with the Christ in Majesty illumination on folio 796 verso. The New Testament begins with a new quire, the one hundred and first, which like most of the others is a quaternion. It opens with Jerome’s Novum Opus on folio 797 recto and verso. Then follow the seven folia of canon tables from 798 recto through 801 recto. The verso of folio 801 is blank. Then follows the Plures fuisse on 802 recto and verso, the preface to the Gospel of Matthew filling the remaining space on 802 verso and continuing onto folio 803 recto with the Chapter list of Matthew also on 803 recto and concluding on 804 verso where they complete the quire.
9The presence of canon tables in the Codex Amiatinus is not surprising. By c.700 we would certainly expect to find them in an Insular manuscript, especially in one produced at a monastic foundation that was such a staunch supporter of the Roman church9. Canon tables already had a long history, their conception attributed to Eusebius of Caesarea in the early fourth century who explained their usage in his now famous letter to Carpianus. Designed to facilitate the study of concordant textual passages in the Gospels, the canon tables were an offshoot of the earlier and overly cumbersome Ammonian system, which placed parallel textual passages from the other Gospels alongside the text of Matthew. While it did provide a method for cross-referencing it also significantly interrupted the train of thought in the reading of the other three Gospels10.
10Eusebius devised an ingenious solution by adding ten tables of concordance before the text of the Gospels enabling cross-referencing without interfering with the continuity of the text. In Canon 1 four vertical columns list the numbers of the section that are common to all four Gospels. The remaining canons contain the following sets of parallel passages: Canon 2, the sections for Matthew, Mark and Luke; Canon 3, the sections for Matthew, Luke and John; Canon 4, the sections of Matthew, Mark and John; Canon 5, the sections for Matthew and Luke; Canon 6, the sections for Matthew and Mark; Canon7, the sections for Matthew and John; Canon 8, the sections for Luke and Mark; Canon 9, the sections for Luke and John; and Canon 10, the sections of the four Gospels for which there are no parallels.
11The effective and efficient Eusebian tables began to be widely used, particularly, after their adoption by Saint Jerome in his translation of the Gospel texts into the Latin Vulgate in the late fourth century. In fact, both Eusebius and Jerome discussed their efforts in letters that were often included in the preliminaries to the canon tables11. We would anticipate that both Eusebius’s letter to Carpianus and Jerome’s letter to the Pope, which explains the new work, his Novum Opus or his Vulgate translation, would be present in the Codex Amiatinus12. However, only Jerome’s Novum Opus is included. Does the omission of Eusebius’s letter reflect access to manuscript models or might this reflect an editorial process exercised by the Wearmouth and Jarrow scriptorium? Given that the Codex Amiatinus was in Jerome’s Vulgate, a characteristic greatly valued by the scriptorium, and that his Novum Opus also included a discussion of the canon tables and their use, there may have been a desire to cull material regarded as redundant if there were a desire to economize on space given the size of the Amiatinus, which when eventually completed was 1,033 folia or 2,066 pages. The scriptorium made a conscious choice in selecting for their prototype the pandect Ceolfrith had brought back from Rome that was in the Old Latin13. However, the scriptorium did use the Latin Vulgate for their pandects, and the inclusion of Jerome’s letter emphasized the worth of this process while also explaining the function of the canon tables.
12Close observation of the canon tables makes it clear that their importance was appreciated by the Wearmouth and Jarrow scriptorium and that every aspect of their creation was a high priority14. Not only are the canon tables in the Codex Amiatinus beautiful to behold, but they also function well. Lest we think that all canon tables linked to Insular scriptoria function, we have only to study those of the Book of Kells (Trinity College Library, Dublin, 58), beautiful but useless, and those incomplete canon tables of the Maaseyck Gospels (St. Catherine’s, Maaseyck) indicating confusion on the part of the scribe that led to the abandonment of the project15.
13In contrast to these other Insular manuscripts the canon tables of the Codex Amiatinus function well, in fact, phenomenally well, for there is also the use of an abbreviated system within the Gospel text, a type of scribal shorthand. At some early point in the history of canon tables it became apparent that a step or steps could be saved if the reader did not have to turn back to the canon table proper but instead could use an abbreviated system placed in the margin next to the text. That way when reading the text in one Gospel one could determine what other parallel passages occurred in other Gospels.
14The Eusebian canon tables coupled with the internal abbreviated system offered an amazingly simple and effective method for cross-referencing the Gospel texts and affirm the anonymous author’s statement in the Life of Ceolfrith that the abbot left two of the pandects upon his departure for Rome to be placed in the churchs so that all who wished could read any chapter in the Testament and might readily find what they wanted16. Both of these systems depended on meticulously accurate copying of the numbers in the tables, the numbers in the section within the text and the letters and numbers of the internal abbreviated system. That all of these function so well within the Codex Amiatinus is a resounding affirmation of the skill, accuracy, understanding and priorities of the scriptorium of Wearmouth and Jarrow.
15However, all of these elements are also shared by the canon tables of the other contemporary manuscripts produced in Northumbria, the Lindisfarne Gospels (British Library, London, Cotton Nero D. IV) and Royal 1B. VII, also a Gospel book housed in the British Library and produced at a house linked to Lindisfarne. The strong relationships among the canon tables and the functioning abbreviated systems within each of these three Northumbrian manuscripts speaks to the breadth, depth and phenomenal quality of the Northumbrian scriptoria as well as the linkages among them17.
16What is quite striking is that although these three Northumbrian manuscripts share similar Gospel texts in the Latin Vulgate and similarities in the canon tables, the number of numerals and the functional abbreviated system, they do not share the same number of folia for the canon tables nor do they share the same architectural frameworks. The canon tables of the Amiatinus are on seven folia, those of the Lindisfarne Gospels on sixteen folia and in Royal 1B. VII on twelve folia.
17If we believe that these manuscripts are linked to the same model, and I would argue that they are, it appears that there were certain elements in this Insular context that were deemed immutable – the numbers and the correlation with the text – and others, particularly the number of folia devoted to the canon tables and the art associated with the architectural frameworks that were considered open to modification. If we believe that they shared the same original model we must accept that this model was modified by the Insular scriptoria but that the functionality of the canon tables was deemed the most important component. The other theory would have to assume that there were three different visual models of canon tables on seven, twelve and sixteen folia all available to the Northumbrian scriptoria at the same time sharing the same numerical tables and abbreviated systems, which seems highly unlikely18.
The Canon Tables considered within the broader Mediterranean context
18It is inevitable that our discussion of the canon tables of the Codex Amiatinus leads us to the fascinating, vexing and complex issue of the transmission of sources within the Mediterranean context. I have approached this issue with some trepidation due to the fact that much of the scholarship linked to the Codex Amiatinus has been overshadowed by the desire to have it serve as a window to the Vivarium of Cassiodorus with its now famous lost manuscripts while other lines of inquiry have been overlooked. In particular the Codex Grandior, a pandect in the old Latin, and the Novem Codices, most likely in the Latin Vulgate, are the most popular candidates from Cassiodorus’ sixth-century scriptorium in the search for the model or models for the Amiatinus. However, it is all too easy to ascribe a connection to ephemeral models and recent works such as John Lowden’s study of the Octateuchs have pointed out the potential pitfuls of the recension theory in art19. However, I think that there are some observations that deserve attention regarding these canon tables and offer suggestive evidence of the type of model or models that were accessible to the scriptorium at Wearmouth and Jarrow.
19Observing the architecture framework of the canon tables, Carl Nordenfalk developed a classification system that is identified by the shape of the framework: the n type, the m type, the m and n type and the m and m type. The Lindisfarne Gospels have the m and n type while the Codex Amiatinus and the Royal 1B. VII both have the m and m type not thought to be present in Insular manuscripts. Also unusual for Insular manuscripts is the use of silver and gold leaf in the Amiatinus and in particular its extensive use on the canon table folia. The simple, attenuated, two-dimensional arch and columns are most similar in form and aesthetics to those in Syriac Gospels although these tables are usually on nineteen folia and often have images intermingled with the tables. What can be said is that the visual model used for the canon tables in the Codex Amiatinus is not one that was available or perhaps not selected for aesthetic reasons by other Insular scriptoria, and that the closest surviving visual models stylistically are in the eastern Mediterranean20.
20The distribution of Amiatine canon tables over seven folia also points to earlier models in Greek. This led Tischendorf to characterize them as the smaller Greek series21. If we also look at the distribution of the each of the tables over the seven folia and compare them to Nordenfalk’s reconstructed Greek series, which he believed to have its origin in Constantinople, there is more to be said on the degree of the close relationship of the Amiatine tables to those of the Greek. If we begin to look at the number of numerals in the canon tables we see some discrepancies between those in the Amiatinus, those in the Greek series and those in the Vulgate. However, if we concentrate on the distribution of the canon tables over the seven folia we see a match between those of the Codex Amiatinus and the Greek series which again places us in the east and also reinforces the link with the Old Latin versions.
21What we appear to have in the Amiatinus is a kind of hybridization of a Latin Vulgate text with canon tables pointing to a model or models affiliated with the Greek series but with modifications. The effective and integrated relationship of the Amiatine canon tables and the abbreviated internal system within the Gospel text, suggests that the model was unified and functional in terms of text and canon tables. That model was also likely produced at a scriptorium that, like Wearmouth and Jarrow, placed a high priority on the canon tables, their visual appeal, their function and their actual use as tables of concordance. The scriptoriun producing the model would also have had strong eastern affiliations, likely some connection to Constantinople or manuscripts produced there, and one that was prior to or nearer to that critical, transitional period when Jerome translated the New Testament into the Latin Vulgate and it became the preferred text.
22The shared distribution of the Amiatine canon with those of the Greek series confirms Bede’s statement is his Life of the Holy Abbots of the Monastery of Wearmouth and Jarrow that Ceolfrith added three pandects of the new translation to the single copy of the old he had brought back from Rome22. The distribution of the canon tables on seven folia indicates that not only was this pandect present at the scriptorium of Wearmouth and Jarrow but it also served as a visual model for the canon tables.
23How many pandects were circulating at this time of such a rare book-type? There were few given our references to pandects in texts, the paucity of surviving pandects and most of all our understanding of the phenomenal efforts that required to produce such a volume. A pandect would have been a volume suitable for a stable and wealthy monastery that prioritized engendering a high degree of learning among its constituents23.
24It is tempting to conclude that the pandect at Wearmouth and Jarrow was from the scriptorium of Cassiodorus at Vivarium in southern Italy. We know that he spent time in Constantinople after leaving Ravenna and prior to going to southern Italy, and his life experience as a scholar and a statesman bridges east and west. Most of all is it tempting to confirm the presence of the most famous and elusive of the Cassiodorean manuscripts, the now lost pandect in the Old Latin, the Codex Grandior. This hypothesis would rule out the arguments supporting the presence of another Cassiodorean manuscripts, the Novem Codices, believed to be in the Latin Vulgate and having a set of canon tables that Cassiodorus refers to in his Institutiones. Or are there other possible candidates perhaps not yet identified due to the lack of surviving evidence or lack of exploration of evidence? In the meantime, the window to the Vivarium seems to be a little easier to see through from my present perspective.
Page distribution and Codicological Factor
Of Pages of Canon Tables | Size of folia | Text area | |
Ms Amiatino I | 4 | 505 x 340 mm | 36-375x260 mm |
Cotton Nero D.IV (Lindisfame Gospels) | 16 | 340 x 245 mm | 235 x 190 mm |
Royal IB.VII | 12 | 285 x 215 mm | 235 x 170 mm |
Total Numerical Citations According to Distribution in Canon Tables
Canon | Ms | Amiatino 1 | Cotton Nero D.IV | Royal 1B.VII | Stuttgart Vulgate * | Greek Series | |||
I | Mt | Mk | Lk | Jo | 284 numbers | 284 numbers | 284 numbers | 288 numbers | 296 numbers |
II | Mt | Mk | Lk | 327 numbers | 327 numbers | 327 numbers | 327 numbers | 333 numbers | |
III | Mt | Lk | Jo | 66 numbers | 66 numbers | 66 numbers | 66 numbers | 66 numbers | |
IV | Mt | Mk | Jo | 78 numbers | 78 numbers | 78 numbers | 75 numbers | 75 numbers | |
V | Mt | Lk | 166 numbers | 166 numbers | 166 numbers | 168 numbers | 164 numbers | ||
VI | Mt | Mk | 96 numbers | 96 numbers | 96 numbers | 96 numbers | 94 numbers | ||
VII | Mt | Jo | 14 numbers | 14 numbers | 14 numbers | 14 numbers | 14 numbers | ||
VIII | Mk | Lk | 26 numbers | 26 numbers | 26 numbers | 26 numbers | 26 numbers | ||
IX | Lk | Jo | 42 numbers | 42 numbers | 42 numbers | 42 numbers | 42 numbers | ||
X | Mt | 62 numbers | 62 numbers | 62 numbers | 62 numbers | 62 numbers | |||
X | Mk | 19 numbers | 19 numbers | 20 numbers | 19 numbers | 19 numbers | |||
X | Lk | 72 numbers | 72 numbers | 72 numbers | 72 numbers | 72 numbers | |||
X | Jo | 96 numbers | 96 numbers | 95 numbers | 96 numbers | 96 numbers |
Comparison of the Distribution of Canon Tables
Canon Page | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
MS Amiatino 1 | I, I | II, II | II, III | IV, V | V, VI, VI, VII | VIII, Mt Mk IX X(2) X(1) | Lk Jo X(3) X(3) | |
Reconstructed Greek Series * | I, I | II, II | II, III | IV, V | V, VI, VI, VII | VIII, IX Mt Mk XX | Lk Jo XX | |
Mt. Athos Lavra A.23 | I | I | II, II | II, III | IV V VI | VI, VII, IX, VIII | Mt Mk Lk X(2) X(1) X(3) | Jo X(4) |
*See C. Nordenfalk, «The Apostolic Canon Tables », Gazette des Beaux Arts, 6:62, July-August 1963, p.17-34. See especially p.21 for the author’s reconstruction of the seven-page ‘smaller’ Greek series whoch partially survived in a VIIth century fragment (London, British Museum, Add. MS 5111). He attributes this manuscript most likely to Constantinople. The distribution of the canon tables over the seven pages matches the distribution of the canon tables in the Codex Amiatinus Please note in Nordenfalk’s reconstructed seven-page series he does not indicate the placement of the tables on the page nor the number of columns in Canon X for each Gospel.
Canon Page | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 |
Cotton Nero D.IV | I | I | I | II | II | II | II | II I | IV | V | V | VI | VII, VIII | IX, X Mt | X Mk, Lk | X, Jo |
Royal 1.B.VII | I | I | II | II | II | II I | I V | V | VI, VII | VIII, IX, X Mt | X Mk X Lk | X Jo |
Notes de bas de page
1 E.A.Lowe, CLA, 3, 299. The additional folia are currently all housed in the British Library in London, as Add. MS 37777, a single folio, Add. MS 45025, consisting of ten folia and fragments of an eleventh, and Loan MS 81, another single folio.
2 Rosemary Cramp has written numerous articles on her excavations. See in particular «Excavations at the Saxon Monastic Sites of Wearmouth and Jarrow», Medieval Archaeology, 13, 1969, p.21-66.
3 D.Wright, «Some Notes on English Uncial », Traditio, 17, 1961, p.442-456. According to Wright the seventh scribe begins his work on folio 797 recto with the Novum Opus, continues writing through all four of the Gospel texts and concludes his work in the Amiatinus on folio 934 verso at the end of the Book of Acts. Wright does not identify this scribe’s hand in any other text.
4 P.Meyvaert, «Bede, Cassiodorus and the Codex Amiatinus», Speculum, 1996.
5 Rupert L. S.Bruce-Mitford’s 1967 Jarrow lecture, «The Art of the Codex Amiatinus », includes a discussion of the canon tables on pages 19-23. He describes the art of the canon tables, his belief in their relationship to those of Lindisfarne Gospels which is based on visual evidence, and concludes by stating that the Amiatine canon tables are clearly linked to Cassiodorus. Bruce-Mitford also discussed the relationship of the Amiatine canon tables to those of the Lindisfarne Gospels in Part IV, Decoration and Miniatures, p.174-185, text volume accompanying the facsimile of Lindisfarne Gospels, Evangelorium Quattuor Lindisfarnensis edited by T. J. Brown.
6 C.L.NeumandeVegvar, The Northumbrian Renaissance, London and Toronto, Associated University Presses, 1987, p.278-280. The author reviews the debate regarding the possible models of the Lindisfarne Gospels citing John Chapman’s and Pierre Courcelle’s belief that the Insular Vulgate stemma, headed by the Codex Amiatinus and shared by the Lindisfarne Gospels and Royal 1B. VII, was derived from the seventh volume of Cassiodorus’s Novem Codices. T. J. Brown postulated a non-Cassiodorean Neapolitan model. Neuman de Vegvar explains that Brown uses the differing number of folia devoted to the canon tables in each of these manuscripts as his reason for arguing a Neapolitan source and proposed that the Amiatinus used a seven pages system for compaction and it came from a Cassiodorean source.
7 P. McGurk, «The disposition of numbers in the latin Eusebian canon table », Philologia Sacra, vol.I, Freiburg, Verlag Herder, 1993, p.242-258.
8 C.Nordenfalk, Die spätantiken Kanontafeln, Göteberg, O. Isacson, 1938.
9 The presence of canon tables in Insular manuscripts is generally thought to be a direct result of the Roman mission. It is often noted that no surviving early manuscripts produced at scriptoria linked to the Celtic church contain canon tables. P. McGurk mentions on page 254 that several manuscripts from the Echternach scriptorium may have Irish connections and could then provide evidence, in addition to the poem of Aileran on the canon tables, of arcaded tables in Ireland.
10 The Ammonian system is attributed to Ammonius of Alexandria. It divided each Gospel into a series of consecutively numbered sections – Matthew has 355, Mark 233, Luke 342 and John 232.
11 These so-called preliminaries were regarded as aids to the intelligent reading of the Gospels and the use of the canon tables. However, their codicological placement is not fixed and they sometimes precede or may follow the canon tables.
12 In fact, both Eusebius’s letter to Carpianus and Jerome’s Novum Opus are in the Lindisfarne Gospels (Cotton Nero D, IV), a manuscript closely linked to the Codex Amiatinus and sharing a similar Vulgate text. In the Lindisfarne Gospels the so-called preliminaries are all placed before the canon tables in the following order: first is the Novum Opus, then the Plures fuisse and then Eusebius’s letter to Carpianus. The production of this Gospel book is attributed to another Northumbrian scriptorium at Lindisfarne approximately forty miles away from Wearmouth and Jarrow. Lindisfarne was a Celtic foundation linked to Iona which adopted the precepts and practices of the Roman Church and was reconsecrated as Saint Peter’s by Archbishop Theodore in 690 on his visit there.
13 Bede clarifies that it was Ceolfrith and not Benedict Biscop who brought back the Old Latin pandect from Rome. See Historia abbatum auctore Bedae edited by C.Plummer. P.Meyvaert emphasizes this in his 1996 article and makes a strong case that the pandect was likely the Codex Grandior from the scriptorium, at Vivarium, of Cassiodorus.
14 The importance of the canon tables was no doubt largely due to their affiliation to the text of the Gospels, the books of the New Testament telling the life of Christ, and the importance of the Gospels is attested by the presence of images from the texts at Wearmouth and Jarrow. See G.Henderson, «Bede and the Visual Arts», Studies in English Bible Illustration, The Pindar Press, 1985. It should also be noted that illuminated folia are rather few in number in the massive Codex Amiatinus with its 1,033 folia or 2,066 pages. Illuminated and decorated folia occur in the first gathering with full-folio miniature of Ezra and the double folia Tabernacle, the Christ in Majesty on folio 796 verso and the seven folia of canon tables.
15 These two manuscripts were likely produced at Iona and Echternach respectively. Françoise Henry discussed the canon tables of the Books of Kells in a text of the same name and discussed some of the numerous problems in the non-functional tables with no corresponding numbers within the Gospel text. N.Netzer in Cultural Interplay in the Eighth Century, Cambridge University Press, 1994, artfully reconstructs the confusion of the scribe in copying the canon tables, the fact that as done the folia could not be incorporated into the Gospels and, therefore, the abandonment of the project. If the canon tables had been understood and valued in that scriptorium, it is likely that the mistake would not have been made or, if made, corrected and the canon table project continued.
16 HAB, 20. In the Historia abbatum auctore anonymo, also edited by C.Plummer, the desire of study and the necessity of the availability of the Bible text is emphasized. The canon tables and the abbreviated system would have greatly facilitated this in the Gospel books of the New Testament.
17 My study of the canon tables of these three manuscripts, the Codex Amiatinus, the Lindisfarne Gospels and Royal 1B. VII indicates that in addition to sharing a similar Vulgate text there are also many similarities among their canon tables and abbreviated systems. However, there are distinctive differences in the art, mainly the architectural arcades of the tables, and the number of folia on which the canon tables are distributed. Canon tables in the Codex Amiatinus are on seven folia, those in the Lindisfarne Gospels on sixteen folia and those in Royal 1B. VII on twelve folia. It is also important to note that the number of folia used for the canon tables does not correlate to the size of the folia. The Amiatinus folia are 505x340 mm with a text area of 360-375x260 mm, the Lindisfarne Gospels are 340x245 mm with a text area of 235x190 mm and the folia of Royal 1 B.VII are 285x 215 mm (some have been trimmed due to damage) and the text area is 235x170 mm.
18 McGurk mentions this and believes that it is highly improbable. R.L.S.Bruce-Mitford believed that the Lindisfarne canon tables were based on the same model as the Codex Amiatinus with modifications due to the potential space available when producing a manuscript that only contained the four Gospels versus the entire Bible and a difference in the aesthetics of the monasteries.
19 JohnLowden, The Octateuchs. A Study in Byzantine Manuscript Illumination, Pennsylvania Park: Pennsylvania State University, 1992. The author presents a strong case for rescinding the confidence once placed in the applicability to the study of manuscript illumination of the recension, theory originally coming from philological study of text and its use to determine the relationships of texts.
20 B.ApelianBeall, «Interchangeable Parts: Interchangeable Art: A Case Study of Three Related Northumbrian Manuscripts». This currently unpublished study of the canon tables of the three Northumbrian manuscripts was presented at the 35th International Congress on Medieval Studies in Kalamazoo, Michigan in May 2000.
21 C.Tischendorf, Novum Testamentum Latine Interprete Hieronymo, Avenarius et Mendelsohn, 1850, identifies the tables in the Amiatinus as the smaller Greek series that occurs on seven folia. Although the distribution is the same of the smaller Greek series the number of numbers is not. See appendices for more on this. It has also been suggested that the reason for the canon tables being placed on seven folia in the Codex Amiatinus was the large size of the folia. However, this argument does not consistently work with the three Northumbrian manuscripts in my case study paper referred to above. In the Amiatinus the canon tables are on seven folia and the dimensions of the folia are 505x340 mm with a text area of 360-375x260 mm. In the Lindisfarne Gospels the canon tables are on sixteen folia with dimensions of 340x245 mm and a text area of 235x190 mm and in Royal the tables are on twelve folia measuring 285 x 215 mm (many folia have been trimmed) with a text area of 235 x 170 mm. If we apply the earlier argument that larger folia means fewer folia used for the canon tables it is not consistent to the evidence of these three closely related Northumbrian manuscripts.
22 HAB, 15. See note 13.
23 I would like to extend my sincere thanks to François Dolbeau of the École Pratique des Hautes Études et Comité Du Cange, Paris for his insightful questions and comments on the nature of the distribution of the canon tables which continue to deserve on-going reflection.
Auteur
Assumption College, Worcester, MA
Le texte seul est utilisable sous licence Licence OpenEdition Books. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.
Se vêtir à la cour en Europe 1400-1815
Cultures matérielles, cultures visuelles du costume dans les cours européennes
Isabelle Paresys et Natacha Coquery (dir.)
2011
Les lettrés de la République
Les enseignants de la Faculté des Lettres de Douai puis Lille sous la Troisième République (1870-1940)
Jean-François Condette
2006
Héros militaire, culture et société (XIXe-XXe siècles)
Claude d'Abzac-Épezy et Jean Martinant de Préneuf (dir.)
2012
Archives, archivistes, archivistique dans l'Europe du Nord-Ouest du Moyen Âge à nos jours
Entre gouvernance et mémoire
Martine Aubry, Isabelle Chave et Vincent Doom (dir.)
2007
Les Cardinaux de la Renaissance et la modernité artistique
Frédérique Lemerle, Yves Pauwels et Gennaro Toscano (dir.)
2009
Élites et sociabilité au XIXe siècle
Héritages, identités
Hervé Leuwers, Jean-Paul Barrière et Bernard Lefebvre (dir.)
2001
Bède le Vénérable
Entre tradition et postérité
Stéphane Lebecq, Michel Perrin et Olivier Szerwiniak (dir.)
2005
De Georges Clemenceau à Jacques Chirac : l'état et la pratique de la Loi de Séparation
Robert Vandenbussche (dir.)
2008
4000 bourgeois de Lille au xive siècle
Le premier Registre aux Bourgeois (1291-1355). Édition et commentaire
Martine Aubry
1999
Un ministre artésien dans la crise du 16 mai
La correspondance entre Auguste et Lucie Paris (16 mai - 23 novembre 1877)
Jean-Marc Guislin
2002