Bede’s Scientific Works in the Carolingian Age
p. 247-259
Résumé
Depuis quelques années, on commence à reconnaître que les oeuvres scientifiques de Bède furent les mieux connues de ses écrits pendant la période carolingienne. Néanmoins, l’étude de l’influence du De temporibus et surtout du De natura rerum et du De temporum ratione sur la culture intellectuelle des Carolingiens reste à faire. Cet essai fait le bilan des études bédiennes, surtout celles de Charles W. Jones et de Frances R. Lipp, et offre quelques précisions sur les manuscrits et les gloses des oeuvres scientifiques de Bède. L’attribution au moine anglo-saxon Byrhtferth de Ramsey des gloses publiées sous son nom par Herwagen (1583), attribution contestée par Jones (1938, 1939, 1977) mais récemment soutenue par Michael Gorman (1996, 2001), est réexaminée.
Texte intégral
1The title of this essay calls to mind the pioneering work of Charles W. Jones (1905-1989) whose many years of fundamental manuscript research mapped the terrain of Carolingian Bedan studies1. Jones appreciated the fact that the Carolingian age knew Bede the scientific writer better than Bede the historian2. The accessus an unknown late ninth-century master provided to Bede’s De arte metrica is instructive. An obvious admirer of the Northumbrian master, the Carolingian teacher knew surprisingly little about him. Bede was “studiossimus in scripturis, praecipue in arte calcularia” and that was about it3. This is not to deny the importance in the Carolingian age of Bede’s hagiography, exegesis, or especially of the Historia ecclesiastica gentis Anglorum, about whose influence we are beginning to learn much more4. But numbers of surviving manuscripts reflect readership and interest and tell their own story. According to Jones’s census, 240 whole or partial manuscripts of DTR survive, the bulk of them originating in the ninth, tenth, and eleventh centuries. The Historia ecclesiastica survives in some 160 manuscripts from the eighth to the sixteenth centuries, with the “boom” period for copying Bede’s history located in the twelfth century5. These numerical and chronological comparisons underscore the impression created by the accessus to De arte metrica: Bede was known in the Carolingian age primarily as an expert “in arte calcularia, de qua scripsit pulcherrimum et utilem librum quem titulauit De temporibus siue De ratione temporum”6.
2Modern scholars along with ninth-century masters have also begun to appreciate Bede’s scientific works. Wesley M. Stevens, the most perceptive modern guide to Bede’s scientific achievement, in his 1985 Jarrow Lecture underscored just how much Bede used observation and reason so that “he not only could see more than others but also could explain well what no one had explained before him”7. The publication of Faith Wallis’s important introduction, commentary, and English translation of DTR will undoubtedly spur wider interest in Bede’s science. Her elegant translation of Bede’s technical vocabulary and the accompanying notes and comments offer the modern reader a “short course” not only to Bede, but to the early medieval computistical tradition as well8. An older work, newly discovered, also contributes to a more precise appreciation of the influence of Bede’s scientific works on the Carolingian age. Frances R. Lipp’s 1961 doctoral dissertation, The Carolingian Commentaries on Bede’s De Natura Rerum, studied nine ninth-century commentaries on DNR9. Six commentaries survive in seven Carolingian manuscripts10. Johann Herwagen, Bede’s sixteenth-century editor, published the three other commentaries from manuscripts now apparently lost. These include glosses he dubbed Incerti auctoris glossae and glosses on both DNR and DTR that he attributed to Byrhtferth of Ramsey (c.970-c.1020), an Anglo-Saxon monk. The Incertus auctor glosses mix scientific explanations with religious and cite Horace, Hyginus, Ambrose, Augustine, Isidore of Seville, Clemens Romanus (Pseudo-Clement) along with the DNR commentary Lipp published in the CCSL edition of Bede’s work11.
3What is required now to fully appreciate the influence of Bede’s scientific work on the Carolingian age is a master study based on the manuscripts that will untangle the attributions of earlier editors and establish the relationships among sets of glosses and commentaries. The great merit of Lipp’s work was to provide a promising, manuscript-based methodology and an important caveat: only by comparing glosses from different Carolingian manuscripts can we sort out how Carolingian masters appropriated Bede’s work. Since the eventual master study must begin with the CCSL editions of DNR and DTR, it will be useful to clarify in the light of recent research some of the statements about the glosses made in those editions.
4When Charles Jones sketched the school context for Bede glossing in 1976, he rightly connected the important Bede glosses in the ninth-century manuscript, MS Berlin, Staatsbibliothek zu Berlin-Preussischer Kulturbesitz, 130 (Phillipps 1832), with the cathedral school of Laon, even though when the glosses were published ascribed them to Metz, the manuscript’s second home12. Jones’s comments in the introduction to the DNR edition were brief, but prescient. He summarized the discoveries of Frances Lipp, reporting her estimate that the DNR commentary in the Berlin manuscript at 6,850 words comes to 1.5 times the length of the text itself. He also suggested that “almost certainly a scholasticus (or two) of Laon composed them [the glosses] at some time between 850 and 873, and our codex, Berlin 130, may well have been copied at Laon”13. Jones’s prefatory comments in the DTR edition were much fuller and included five pages devoted to the glosses and commentary14. Here, Jones made several important statements regarding the Berlin manuscript and its glosses:
“All the texts of Bede contained in this codex were written A.D. 873/874. The annus praesens is incorporated five times in the glosses for DTR”15.
The DNR commentary, following the determination of Frances Lipp, is the work of a single teacher, heavily influenced by Martianus Capella.
Several indices (use of Tironian notes, use of Greek, familiarity with computus, use of sources) suggested to Jones that the DTR glosses might be by a different teacher, perhaps Martin of Laon, nevertheless he concluded that “Despite the apparent divergences, I am inclined to believe that the commentaries on DNR and DTR are products of a single mind.”16
Jones then turned to the Berlin manuscript, which with its fragment, MS Berlin, Staatsbibliothek zu Berlin-Preussischer Kulturbesitz, 129 (Phillipps 1830), once composed a single codex: “It seems reasonable to surmise that Berlin MSS 129 and 130, or their immediate exemplar, were written at Laon in the year 873/4 and that DNR and DTR commentaries were composed by one scholasticus, or possibly two, then or not long before that date.”17
5Jones’s suggestions about the Berlin manuscript, although remarkably accurate in general terms, can be brought into sharper focus. First, the manuscript’s Bede texts and the accompanying glosses were indeed copied at Laon. The principal hand of the Laon Annals in Berlin 129 is that of Martin Hiberniensis (819-875), the Irish magister Laudunensis. Martin not only copied historical entries alongside Bede’s 19-year tables, he also assisted the manuscript’s principal scribe in copying the manuscript18. On fol. 27v, when copying DTR 15, with its unusual terminology of the English months, the scribe’s eye skipped a line from the word mensis following Halegmonath to the same word following the next month, Blodmonath. Martin caught the error and supplied the missing text from the exemplar19. Of course, that the manuscript and its glosses were copied at Laon does not necessarily mean that the glosses were composed at Laon. They could have been composed anywhere and copied into the Berlin manuscript when its Bede texts were copied. Here a comparison of the format of the DNR and DTR glosses is instructive.
6The DNR glosses are carefully copied between the lines and in the margins of the manuscript by the same hand that copied Bede’s text. The many longer glosses are keyed to their lemmata by letters of the alphabet, as Lipp’s edition in CCSL 123A clearly indicates. The marginal glosses required three alphabetical cycles, A through Z, and the beginning of a fourth cycle to present the longer glosses. Two of the four cycles are incomplete. The third cycle lacks gloss I and ends with gloss N, while the fourth cycle begins with B, omits E and F, and ends with H. While it might be tempting to attribute such anomalies to the scribe, the scribe of the Berlin manuscript was careful. He might have missed one or two glosses, but significant gaps in the cycle suggest that the source he copied was defective. This means that the DNR glosses were carefully copied into Berlin 130 from some source and not likely composed at Laon.
7In contrast, the appearance of the DTR glosses is totally different. They are not copied by the hand of the principal scribe, but by Martin Hiberniensis himself. His explanations are not carefully laid out on the page as are the DNR glosses, but wind between lines and in margins and among themselves as an original composition, entered in the Berlin manuscript for the first time. Jones’s edition used parentheses and the abbreviation Gl. for Glosa to indicate Martin’s interlinear glosses to his own marginal glosses, as in the note to DTR 8,54: sidus:
Stella (Gl.: id est Mercurius) uero, quae Stiluos grece nuncupatur, quam ei (Gl.: id est Mercurio) pagani asscribunt ex quo (Gl.: id est Mercurio) etiam diei nomen inuenerunt, tanto celerior planetis omnibus [ed. mnibus] currit ut in VIIa die (Gl.: id est in quantum ad solem) suos permeat circulos, quod Saturnus (Gl.: sc. secundum Martianum) XXVIII annis et Iupiter XII possunt20.
8This “layered” pattern of glosses here and elsewhere provides clear evidence of a teacher teaching, of a master who wanted to make sure that the reasoning behind the doctrine was clear. But to whom? It is unlikely that a book such as the Berlin Bede, the master’s own copy, would be passed around to students. It also seems unlikely that the master would need to copy out explanations of his own explanations, e.g., that ei and quo referred to Mercury. Rather, Martin prepared his glosses for masters who would follow him. He may well have had his successor in mind as he copied the glosses in 873 and 874, the dates that he used for computistical examples. Martin undoubtedly taught all of Bede’s DTR, his editorial work in the copying of the Berlin manuscript proves that, but his DTR glosses end abruptly on folio 47v, in chapter 51 of Bede’s work21. Martin’s glossing activity in the manuscript appears to have ended with his death in 875.
9Martin’s glosses are a remarkable source for the study of Bede’s scientific works in the Carolingian age, not least of all because in an intellectual genre characterized by anonymity and chronological incertitude, they can be localized to a time and place and attributed to a specific master. The only other comparably specific set of glosses are the notes Heiric of Auxerre added to his own Bede manuscript, MS Melk, Stiftsbibliothek 412 (370 G 32). But his glosses, many in Tironian notes, have not yet been published, much less studied systematically22. Although Martin’s death precluded the transcription of a full set of glosses on Bede, his glosses did enjoy posthumous dissemination and unexpected influence.
10Martin’s Bede manuscript remained at Laon for at least a half century after his death where it continued to provide the basis of instruction in Bede’s scientific works. Adelelm of Laon, dean and treasurer of the cathedral before he became bishop (921-930), made personal entries in the manuscript’s Laon Annals for 892, 903, and 911, and copied a form letter based on a letter sent to his predecessor, Bishop Dido of Laon (ca. 882-895) onto the manuscript’s last leaf23. The letter concerns the transfer of a priest from the church of St. Mary at Compiègne to Dido’s diocese.
11Compiègne had very strong connections to the school at Laon and may have been a second site for the dissemination of Martin’s instruction on Bede’s scientific works. In 865, King Charles the Bald established a community of canons at Compiègne and appointed Hedenulf, a Laon priest, to guide them. It appears that at this time, Manno, the Laon scolasticus born in 843, began his teaching career at Compiègne. By the time he left Compiègne in 877, right after Charles’s death, his alumni included three future bishops: Mancio of Châlons (893-908), Radbod of Utrecht (899-917), and Stephen of Liège (901-920). A ninth-century Bede manuscript that contains a second notice of Manno’s birth and a reference to his ordination as priest in 876, MS Trier, Stadtbibliothek 2500, might have accompanied Manno from Laon to his new teaching post. The Bede texts in the Trier manuscript were copied directly on the texts in Martin’s Berlin manuscript. That only a handful of his glosses, probably remembered from oral instruction, appear in the margins of the Trier manuscript suggests that it left Laon before 873/874 when Martin copied his glosses into the Berlin Bede24.
12Martin’s glosses made a much more significant contribution to a remarkable set of glosses attributed to Byrhtferth of Ramsey (c.970-c.1020). Byrhtferth was an accomplished computist who learned much from Abbo of Fleury (c.950-1004). Considerable parts of Byrhtferth’s computus and his Enchiridion can be traced to the tutelage and books Abbo brought to Ramsey when he taught there in 985-98725. In 1563 when Herwagen published his edition of Bede’s works, he also published along with the texts of DNR and DTR a series of glosses that he attributed to Byrhtferth. The glosses are most easily accessible today at the foot of the columns of volume 90 of Migne’s Patrologia latina where they share space with a Vetus commentarius originally published in 1537 by Noviomagus (John Bronchorst), Noviomagus’s own Scholia, and the Incerti auctoris glossae also printed by Herwagen in his 1563 edition26. In 1938 and again in 1939 Charles W. Jones cast doubt on Herwagen’s attribution of the glosses to Byrhtferth. Jones preferred to call the glosses the “Auxerre glosses” because they mentioned Haimo and Remigius of Auxerre and seemed to depend to a certain extent on Heiric of Auxerre’s glosses in his Melk Bede27. If, in questioning Byrhtferth’s connection to the glosses Herwagen published, Jones made an essentially negative contribution to scholarship, his study of the sources and interests of the compiler of the glosses provided the first detailed analysis of the so-called Byrhtferth glosses (hereafter “B glosses”).
13In 1996 Michael M. Gorman re-opened the case for attributing the B glosses to Byrhtferth by explaining “how many of the conclusions reached by Jones are incorrect”28. However, Gorman’s reading of Jones’s 1938 article and its essential republication in Bedae Pseudepigraphica in 1939 misstates several important points Jones made. Jones’s sixteen-page study certainly did not “present the full-case against Byrhtferth’s authorship”29. Rather, Jones concluded only “that Hervagius’s edition of Bede’s computistic works is not alone sufficient authority for deciding disputed matters”30. Far from “imagining” that Herwagen “invented” the attribution of the glosses to Byrhtferth, Jones devoted more than two pages of his brief article to an attempt to discover “Where did Hervagius get the name ‘Bridefertus’?”31. Jones’s footnote 12 provides the reason Gorman sought for Jones’s statement that Herwagen had not reprinted Bede’s DTR from an earlier edition: “My statements are based on collation of texts”32. And Jones did not remain “silent on the problem of the Byrhtferth glosses” almost forty years later when he published the CCSL edition of DTR, but reiterated his belief that Herwagen’s glosses were “wrongly ascribed to Byrthferth of Ramsey”33.
14Jones was the first scholar to appreciate the relationship between the B glosses and the glosses Martin Hiberniensis copied in his Bede manuscript. In 1938 he wrote that duplication between the two sets of glosses was “so close as to be almost complete.” A year later in Bedae Pseudepigraphica, he wrote that the duplication was “constant, but by no means complete”34. Clearly, as Gorman pointed out, further study convinced Jones to modify his original analysis, but Jones did not go so far as to consider “Heiric’s glosses in the Melk manuscript to be a more important source for the Byrhtferth glosses than Martin’s glosses in the Berlin manuscript”35. In fact, Jones’s comparison suggested just the opposite and is of such significance for the study of Bede’s scientific works in the Carolingian age that it deserves full citation. After stating that the majority of glosses in Heiric of Auxerre’s Bede manuscript “are duplicated in B,” (Jones’s B= Herwagen’s Byrhtferth glosses), Jones continued:
In a careful check of thirty pages of the Melk MS. [Heiric’s Bede] I have concluded that any glosses not included in B have been eliminated by chance vicissitudes and not deliberately. In other words, B (which has many times the number of words contained in the Melk glosses) is a transcript increased in size many times by extensive additions from other sources. One of these other sources was a member of the family to which the Berlin MS. [Martin’s Bede] belonged; for so far as I can check by comparing photostats of Berlin MS., 130, fols. 17v-22v, against photostats of Melk MS., G 32, pp.62-82, there are no duplications in the glosses. Yet the texts of DTR in the two manuscripts belong to the same general family...36
15The phrase in Jones’s statement presented in bold here was replaced in Gorman’s citation of the statement by ellipses. Yet, the bolded phrase is crucial to understanding the points Jones made: the majority of Heiric’s glosses appear in B; those that do not, dropped out by “chance vicissitudes”; B is a much larger work than Heiric’s glosses since it includes additions from other sources; one of the sources is related to Martin’s Bede manuscript. Jones’s comparison of two sets of photostats led him to conclude that there is no duplication between Heiric’s glosses and Martin’s glosses. Jones’s comparison was not between the B glosses and Martin’s glosses as Gorman thought, but between two sources (Martin and Heiric) of the B glosses. All Jones was claiming here was that Heiric and Martin were not duplicative37. Jones repeated his conclusion about the relationship of the B glosses and those of Martin in unmistakable terms: “So constant is the relation between B and the manuscript [Martin’s Bede] that we must assume that they come from a single, definite body of material, though each editor has made rearrangements, selections, and possible additions”38.
16Jones provided a clue to what that body of material might have looked like when he published as an appendix to Bedae Pseudepigraphica “An Anonymous Commentary on DTR”39. He knew this text from four manuscripts, the earliest of which is Heiric’s copy of Bede where it appears on pp.32-38 of the Melk manuscript. Here the text was written in continuous lines with comments following lemmata from the DTR. As Jones saw, almost all the comments that constitute this text ended up in the B glosses40. But Jones was mistaken when he attributed the commentary to “the glossator of the Melk MS.” (Heiric?)41. The principal scribe of the manuscript copied the anonymous commentary when the manuscript was first written sometime during the first decade of the ninth century. The text as it appears in the Melk Bede certainly is not an original composition. Rather it pre-dates the manuscript since in two places, the scribe skipped over phrases which he later was able to add in the margin42. While some modern scholars have been deceived by the anonymous commentary’s title in the manuscript, De Natura Rerum et Temporum Ratione, into thinking that it was an independent treatise, a ninth-century reader (perhaps the scribe) knew exactly what the text was when he added appropriate chapter references to Bede’s DTR in the margins43. The commentary deserves fuller analysis. Suffice it here to note that its preservation in the Melk manuscript establishes that a tradition of continuous commenting on the DTR apart from the text of Bede’s work was already underway by the end of the eighth century. The commentary may also have served as the Ur-text of the B glosses. As Gorman observed, the manuscript of the B glosses that Herwagen published likely was not a copy of Bede’s text with copious glosses, but a free-standing commentary44, one very much like the commentary Jones published, but much larger and more complex.
17We should not hold our breath waiting for the recovery of Herwagen’s manuscript. What would be most useful most immediately to understand the enterprise of Carolingian Bedan studies is a re-edition of the B glosses along the lines indicated by Michael Gorman and Frances Lipp.Gorman re-edited the B glosses of DTR 5, presenting the text in a format that makes consultation and analysis much more convenient than that of volume 90 of the Patrologia latina. More importantly, Gorman traced the sources of the glosses, both those that were named by the compiler(s) of the glosses and those that went uncredited45. The obvious usefulness to scholarship of such a re-edition would only be enhanced by the kind of detailed manuscript research Frances Lipp brought to her study of the B glosses on DNR.
18Lipp meticulously analyzed nine Carolingian commentaries on DNR, including six unpublished series of glosses “characterized by a conspicuous absence of unity... [that] can sometimes be traced to the fact that a commentary often represents the combined efforts of several teachers”46. Almost all the material Lipp investigated was composed in centers in northern France. In addition to describing and commenting on each set of glosses and its manuscripts, she edited and compared the glosses on DNR 12 and 20 in order to determine relationships among the series of glosses. Her investigation is a milestone in the study of Bede’s DNR and Carolingian culture and deserves to be better known. Here, I can only summarize the most salient points of her fourth chapter on the B glosses47.
19Lipp’s investigation of the glosses Herwagen published led her to conclude that “[Bn], [Bt] and [T] are not three independent commentaries on the DNR, but are rather three versions of a work of compilation, not extant in other texts, which I have called [B]” (192). The compilation combined glosses on DNR from Martin’s Berlin manuscript, from MS Paris, BNF, n.a.l. 1615 (s.9; Auxerre), and from a commentary related to that in MS Vatican, Reg., lat. 644 (s.10; Saint- Gall?). Through these sources, the three versions of B glosses on DNR are related to all the six unpublished commentaries on DNR. Lipp detected the “combined efforts of at least four or five glossators” in the B glosses on DNR (193). The most complete version is Bn, which glosses DNR 1-22 with 131 glosses consisting of some 8600 words48. Although Bt glosses a smaller section of DNR than does Bn, at 5000 words it is almost as long and seven times longer than the chapters of DNR that it glosses. Since Bt is embedded in Herwagen’s B glosses on DTR, Lipp’s identification of the ten DNR glosses nested in the DTR glosses is especially valuable (199-200).
20The Bn and Bt glosses differ only in form. The few glosses that Bt does not share with Bn are quite similar to the glosses the two commentaries share in common. T, Lipp’s siglum for the Incerti auctoris glossae, glosses the prefatory verses and fifteen chapters of DNR. Lipp discovered that this commentary actually consists of two components that she designated as T1 and T2. T1 consists of glosses on the preface and on six chapters of DNR. The 1400 words of these glosses, like Bt, are closely related to Bn. The T2 glosses are very different from T1 and from Bn, Bt, and the commentaries in the manuscripts. The ten glosses that make up the T2 layer of T pursue religious themes and emphasize allegorical interpretations and the greatness of God as explanations for natural phenomena. Lipp did not recognize any of the T2 glosses in Bn, Bt, or in any of the other commentaries she studied.
21Despite differences among them, Lipp concluded that all three versions of B (Bn, Bt, and T1) derived ultimately from a commentary in which the DNR glosses in Martin’s Berlin manuscript were combined with the glosses in MS Paris, BNF, n.a.l. 1615. In the second half of her ChapterIV, she focused on Bn, the most complete of the B commentaries, in order to examine the nature of the process of compilation. Her edition of seven DNR glosses (210-212) nicely complements Michael Gorman’s edition of DTR 5. Where Gorman identified the sources of the glosses in his sample, Lipp identified the layers of compilation from different manuscripts in her sample. Her analysis of the relationships among the layers of compilation is too complex to present here succinctly. Her most important conclusions were:
that T1 represents an early stage in the compilation
that the B glosses (Bn, Bt, and T1) were not sources for the DNR commentaries found in the manuscripts she studied. They “do not represent a source, but rather a compilation which has drawn on several commentaries for material” (217);
the commentary that accounts for most of the Bn glosses is the commentary in Martin’s Berlin manuscript which Lipp subsequently published in 1977 in CCSL 123A, although the B glosses were not copied directly from Martin’s manuscript
many of B’s interlinear glosses come from a text similar to that of MS Paris, BNF, n.a.l. 1615
the B glosses are also related, but less consistently so, to a set of glosses found in MS Vatican, Reg. lat. 644
the commentaries found in the Berlin, Paris, and Vatican manuscripts are the only ones that show consistent correspondence to the versions of B.
22Michael Gorman did not take up the question of Herwagen’s attribution of the DNR glosses to Byrhtferth, but Lipp did49. After laying out Jones’s arguments against assigning the glosses to Byrhtferth, Lipp agreed with Jones that neither the B glosses on DTR or DNR were compiled by Byrhtferth (233). Lipp, however, looked more closely at the DNR glosses than either Jones or Gorman. What she saw is highly interesting. Not only are the DTR and DNR commentaries quite different in their use of sources, content, and organization, she also suggested that the DNR glosses were the work of another and possibly earlier compiler. The DNR glosses, unlike the DTR glosses Jones attributed to Auxerre, contain none of the markers Jones used to claim an Auxerre origin (glosses by Heiric of Auxerre; mentions of Haimo of Auxerre and Remigius of Auxerre; use of Macrobius). Lipp suggested that Remigius of Auxerre “inherited the DTR glosses in a form very much like that of [Bn], and revised and added to them to suit his purposes, but that he has had little or nothing to do with [Bn]” (238). Thus, not only would the B glosses on DNR appear to be earlier than the DTR glosses, they are also less developed50.
23Lipp’s work on the DNR glosses adds weight to Jones’s argument that Byrhtferth of Ramsey had nothing to do with the glosses Herwagen published under his name. Even so, one might still be tempted to assert that “There is no reason why Byrhtferth of Ramsey could not have compiled the glosses attributed to him by Herwagen”51. But the publication of Byrhtferth’s Enchiridion and extracts from his computus in 1995 makes it easier to lay to final rest Byrhtferth’s claim to the glosses Herwagen published in 156352. There is no connection between the B glosses and Byrhtferth’s authentic works. The DTR glosses cite Martianus Capella while Byrhtferth never did53. Byrhtferth cited the computistical work of Helperic and of Abbo of Fleury, but the B glosses do not. Byrhtferth mixed lessons in Anglo-Saxon with lessons in Latin for his pupils, while the B glossators worked exclusively in Latin. The B glosses on DTR 15, “De mensibus Anglorum,” are remarkably terse on a subject of great natural interest and appeal that one would have expected Byrthferth to have developed fully for his Anglo-Saxon audience54. Matters of style also distinguish the B glosses from Byrhtferth’s authentic work. Jones noted that Byrhtferth was much more fulsome in his characterization of Bede than were the B glossators. For Byrhtferth, Bede was a venerable astronomer, most reverend Bede, a venerable computist, the most learned of men, the blessed man and noble scholar, the marvelous teacher, the venerable writer, and the most learned man among the English. The most the B glossators could manage was dominus Beda55. According to his editors, the phrase “libet hic... inserere” in several forms was “a Byrhtferthian cliché”56. Although the last compiler of the B glosses made many internal cross references to his material and to his excerpting, he never used Byrhtferth’s useful locution57.
24Many of these arguments are made from silence, but cumulatively the silence is thunderous. The B glosses have nothing to do with Byrhtferth or with tenth- century England. Can we end on a more positive note? The work of Charles W. Jones, Frances R. Lipp, Wesley M. Stevens, Faith Wallis, and others has shed important new light on the intense study of Bede’s scientific works during the Carolingian age. We now know that a number of masters glossed Bede’s works. DTR was the subject of a continuous commentary independent of Bede’s text as early as the first decade of the ninth century and probably earlier. By the third quarter of the century, that commentary was pulled apart and individual glosses from it served new masters, such as Heiric of Auxerre and Martin Hiberniensis. Others contributed to the enterprise. DNR was glossed independently of DTR, but the DNR glosses were not as often compiled as the richer, more complex DTR glosses. At Auxerre and Laon these streams of glosses were united in extant schoolbooks, the Melk and Berlin Bede manuscripts. By the last quarter of the ninth century, someone, perhaps Remigius of Auxerre, began systematically to organize a summa of Carolingian glosses on DTR. These B glosses testify to the variety of approaches Carolingian teachers and thinkers took toward Bede’s work. Technically precise glosses share space with glosses inspired by religious concerns. Most of all, the B glosses testify to the richness and depth of Carolingian studies of Bede’s scientific works.
Notes de bas de page
1 C.W.Jones, “The ‘Lost’ Sirmond Manuscript of Bede’s ‘Computus’”, English Historical Review, 52, 1937, p.204-19; Id., “The Byrhtferth Glosses,” Medium Ævum, 7, 1938, p.81-97; Id., Bedae Pseudepigraphica: Scientific Writings Falsely Attributed to Bede, Ithaca, Cornell University, 1939; Id., “Bede’s Place in Medieval Schools”, in Famulus Christi: Essays in Commemoration of the Thirteenth Centenary of the Birth of the Venerable Bede, ed. G.Bonner, London, SPCK, 1976, p.261-285. For a complete list of Jones’s publications, see W.M.Stevens, “CharlesW.Jones: A Bibliography”, in Saints, Scholars, and Heroes: Studies in Medieval Culture in Honour of Charles W. Jones, 2 vols., ed. M.H.King and W.M.Stevens, Collegeville, Minn., St. John’s Abbey and University, 1979, vol.1, p.299-309.
2 We owe our modern editions of Bede’s scientific works to Jones. See Bedae Opera de temporibus, Cambridge, Mass., Medieval Academy of America, 1943; De natura rerum liber (=DNR), CCSL 123A, Turnhout, Brepols, 1975; De temporum ratione liber (= DTR), CCSL 123B, Turnhout, Brepols, 1977; De temporibus liber, CCSL 123C, Turnhout, Brepols, 1980.
3 Accessus ad auctorem Bedam, ed. C.W. Jones, CCSL 123C, p.701,4 (p. 700-702).
4 Three papers presented at the May 2002 37th International Congress on Medieval Studies (Kalamazoo, Michigan) opened new perspectives on the influence of Bede’s history in the Carolingian period: P.Kershaw, “The Peregrini Read Bede: Sedulius Scottus, Berne Burgerbibliothek MS 363, and the Historia Ecclesiastica”; A.Thacker, “Narratives of Reform, Patterns of Ordering: The Impact of Bede upon the Carolingian World”; and J.A.Westgard, “Dissemination and Reception of the Historia Ecclesiastica Gentis Anglorum of the Venerable Bede: An Austrian Family of Manuscripts”.
5 See Jones, “Bede’s Place in Medieval Schools”, p.268. For the eighth- and ninth-century manuscripts, see now, W.M.Stevens, “Bede’s Scientific Achievement: The Jarrow Lecture 1985 (revised 1995)”, in W.M.Stevens, Cycles of Time and Scientific Learning in Medieval Europe, Aldershot, Ashgate, 1995, ch. II, 53-59. For the Historia ecclesiastica I am indebted to Westgard, “Dissemination and Reception” (note 4 above).
6 Accessus ad auctorem Bedam (note 3 above), p.701, 4-6. The master also knew Bede’s DNR: “Composuit et libellum De quadrifario opere Dei” (ibid., p.31).
7 “Bede’s Scientific Achievement” (note 5 above), p.44 (p. 19 dans la Jarrow Lecture de 1985).
8 F.Wallis, Bede: The Reckoning of Time, Liverpool, Liverpool University, 1999.
9 The dissertation (1977) was presented to Yale University in 1961. The first extensive use of Lipp’s findings appeared in J.-J.Contreni, “John Scottus and Bede”, in History and Eschatology in John Scottus Eriugena and His Time, ed. J.McEvoy and M.Dunne, Leuven, Leuven University, 2002, p.91-140.
10 MSS. Berlin, Staatsbibliothek zu Berlin-Preussischer Kulturbesitz, 130 (Phillipps 1832); Amiens, BM, 222; Paris, BNF, n.a.l., 1632; Vatican, Reg. lat., 1260; Paris, BNF, n.a.l., 1615; Milan, Ambrosiana, D 48 inf.; Vatican, Reg. lat., 644. The commentary in MS Berlin 130 subsequently was published in the 1975 CCSL edition of Bede’s DNR. MS Trier, Stadtbibliothek, 2500, is another ninth-century witness to this commentary (see Contreni, “John Scottus and Bede”, p.133-134).
11 Herwagen’s texts appeared in volumes I and II of his Opera Venerabilis Bedae, Basel, 1563, and were reprinted in J.-P.Migne, Patrologiae cursus completus... series Latina, vol.90, Paris, 1850. See the excellent study of Herwagen’s edition of Bede by M.Gorman, “The Canon of Bede’s Works and the World of Ps. Bede”, Revue Bénédictine, 111, 2001, p.399-445, in which Herwagen is characterized as “[i] ntent on publishing many inauthentic works [of Bede]” (p.399).
12 Jones, “Bede’s Place in Medieval Schools”, p.273-275; De natura rerum liber, CCSL 123A: “Cura et studio Ch. W. Jones una cum Commentariis et Glossis e codice Mettense (Berlin MS Phill. 1832) cura et studio Fr. Lipp”; De temporum ratione liber, CCSL 123B: “[...]una cum Commentariis et Glossis scriptis A.D. DCCCLXXIII e codice Mettensi (Berlin MS Phill. 1832) cura et studio Ch. W. Jones”.
13 CCSL 123A, p.185.
14 CCSL 123B, p.241-261 (“Glosses and Commentary”= p.257-261).
15 Ibid., p.257. Actually, the dates are valid only for the glosses.
16 Ibid., p.259.
17 Ibid., p.260. What Jones wrote in the penultimate paragraph of his introduction, p.260-261, is confused. The Trier annal of 865 refers to the establishment of canons in Compiègne, not Laon, under the direction of Hedenulf, a Laon priest and later bishop (876-882).
18 See J.-J.Contreni, The Cathedral School of Laon from 850 to 930: Its Manuscripts and Masters, Munich, Arbeo Gesellschaft, 1978, p.46-65; Idem, Codex Laudunensis 468: A Ninth-Century Guide to Virgil, Sedulius, and the Liberal Arts, Turnhout, Brepols, 1984; and, especially, Id., “John Scottus and Bede”, p.111-116.
19 DTR 15, 47-49 (CCSL 123B, p.332). Martin’s addition reads, “[...] mensis sacrorum. Vintirfillith potest dici composito nouo nomine hiemiplenium. Blothmonath [...]” Jones did not collate Martin’s Berlin manuscript for the text of DTR.
20 MS Berlin 130, fol. 22r. CCSL 123B, p.302.
21 CCSL 123B, p.440, 53 (glosses on tertia and atque).
22 See Contreni, “John Scottus and Bede”, p.105-111. There seems to have been little if any relationship between the glosses of the two masters. For images of p.2, 27, 29, 59, 62, 66, and 192 of Heiric’s manuscript, see the “Inventar der Handschriften des Benediktinerstiftes Melk” at http://www.oeaw.ac.at/ksbm/melk/inv1/. I believe that many of the glosses in another Auxerre Bede manuscript, MS Paris, BNF, n.a.l. 1615, are Heiric’s.
23 See V.Rose, Verzeichniss der lateinischen Handschriften der königlichen Bibliothek zu Berlin: Erster Band (Die Meerman-Handschriften des Sir Thomas Phillipps), Berlin, A. Asher, 1893, p.293, for a transcription of the letter which begins (fol. 92v) “[...] doni amantissimo atque reuerentissimo patri et coepiscopo”. Rose thought the salutation referred to Bishop Udo of Trier (1066-1078), but the addressee is Bishop Dido (Didoni) of Laon (ca. 882-895).
24 For Manno and the Trier manuscript, see Contreni, “John Scottus and Bede”, p.131-137. Also, A.-M.Turcan-Verkerk, “Manno de Saint-Oyen dans l’histoire de la transmission des textes”, Revue d’histoire des textes, 29, 1999, p.169-244.
25 See M.Mostert, “Abbo”, in The Blackwell Encyclopaedia of Anglo-Saxon England ed. M.Lapidge et Al., Oxford, Blackwell, 1999, p.3. For Byrhtferth, his Enchiridion, and substantial portions of his computus, see Byrhtferth’s Enchiridion, ed. P.-S.Baker and M.Lapidge, Oxford, Oxford University, 1995.
26 PL 90, c. 187A-278A. For the Incerti auctoris glossae, see V.V.Petroff, “Carolingian School Texts: Glosses from the Circle of John Scottus and Remigius of Auxerre” (in Russian), in Philosophy of Nature in Antiquity and the Middle Ages, ed. P.P.Gaidenko and V.V.Petroff (in Russian), Part 2, Moscow, Russian Academy of Science, Institute of Philosophy, 1999, p.233-292. For a summary of Petroff’s findings, see Contreni, “John Scottus and Bede”, p.100-102.
27 Jones, “Byrhtferth Glosses” (see above, note 1); also, the next year in Bedae Pseudepigraphica, p.21-38. For the “Auxerre glossator” in Jones’s Bedae Opera de temporibus, see p.82, note 4, p.119, note 2, p.146, p.332-333, p.350-351, p.359, p.364, p.367-368, p.372, p.381, p.383, p.386.
28 M.Gorman, “The Glosses on Bede’s De temporum ratione attributed to Byrhtferth of Ramsey”, Anglo-Saxon England, 25, 1996, p.209-32 (quotation from p.210); reprinted in Id., Biblical Commentaries from the Early Middle Ages, Florence, Sismel-Edizioni del Galluzzo, 2002, p.175-198. This judgment was repeated in 2001 in Id., “The Canon of Bede’s Works”, p.429 (see above, note 11).
29 Ibid., p.210.
30 Jones, “Byrhtferth Glosses”, p.97.
31 Gorman, “Glosses”, p.211; Jones, “Byrhtferth Glosses”, p.94-96. Jones suggested that Herwagen concluded from John Bale’s list of Byrhtferth’s works that Byrhtferth was responsible for the glosses. Jones pointed out that early modern bibliographers were often imprecise when reporting titles and incipits of medieval works and found it significant that Herwagen and Bale agreed in their confusion. On p.213 of his essay, Gorman acknowledged Jones’s argument, but thought it conjectural.
32 Jones, ibid., p.82, note 12; Gorman, ibid., p.211: “Jones... offered no reasons for his belief”.
33 Gorman, ibid., p.214; Jones, CCSL 123B, p.257. It is curious that such a careful and critical scholar should misunderstand Jones. In “The Canon of Bede’s Works”, p.426 (see note11 above), Gorman wrote: “Jones did not find a single manuscript that was used by Herwagen, and concluded, ‘Not a single computistical manuscript he used, and he must have used more than one, is known today.’ (p.21 [Bedae Pseudepigraphica]) This assessment seems to me unduly pessimistic. An exclamation point (!) is placed before those items which are known to have been printed from an extant manuscript.” But, Jones wrote “computistical manuscript.” Gorman marked 11 items in his valuable annotated table of contents to Herwagen’s eight volumes with exclamation points (“The Canon of Bede’s Works”, p.427-439), not one of which is a computistical work. Gorman’s own analysis confirmed Jones’ pessimism.
34 Jones, “Byrhtferth Glosses”, p.88; idem, p.27.
35 Gorman, “Glosses”, p.213.
36 Jones, Bedae Pseudepigraphica, p.29.
37 Gorman, “Glosses”, p.213.
38 Jones, Bedae Pseudepigraphica, p.28. Gorman cited this passage immediately after writing: “Jones had evidently come to question his earlier assertions that the Byrhtferth glosses have passages in common with Martin’s glosses in Berlin Phillipps 1832, although this is not completely clear” (“Glosses”, p.214). Jones’s statement becomes clear when one realizes that he did not question his earlier conclusions about the close relationship between the B glosses and Martin’s glosses.
39 Jones, ibid., p.95-102. See p.29-31 for his discussion of the text.
40 He explained why three passages from the text were omitted from the B glosses. Pace M.Gorman (“Glosses”, p.214), the text Jones published extends beyond the preface of the DTR to chapter 18. Also, Jones was aware of the connection between the text he published and the B glosses. He entered apposite references to the B glosses in PL 90 in the margins of his edition.
41 Jones, Bedae Pseudepigraphica, p.29.
42 Ibid., p.100, 5-6 (“xxx habeat... annorum”) and lines 28-30 (“usque tempus... perveniat”). At p.101, 7, Jones skipped after “Agenoris filiam” the manuscript’s “Cretam venit in navi”.
43 The title of the commentary might originally have been the opening words of Bede’s preface to DTR: “De natura rerum et ratione temporum...” (CCSL 123B, p.263, 1). The first B gloss (PL 90, c. 686C-687C: “Non est praetermittendum...”) reproduces the commentary’s first note (Jones, Bedae Pseudepigraphica, p.95-96).
44 Gorman, “Glosses”, p.221-222.
45 Ibid., p.226-231.
46 See above, note 9. Her description of the commentaries appears in the dissertation’s abstract. Lipp later published one set of DNR glosses from Martin Hiberniensis’s Bede manuscript in CCSL 123A, p.192-234. For discussion of these glosses, see Contreni, “John Scottus and Bede”, p.116-123.
47 “Commentaries [Bn], [Bt] and [T]: The Three Versions of [B]”, p.191-268. Bn= the glosses on DNR that Herwagen published and attributed to Byrhtferth. Bt= the portion of the DTR glosses that contains glosses on ten chapters of DNR. T= the Incerti Auctoris Glossae also published by Herwagen. The following paragraphs summarize the findings of Lipp’s ChapterIV. I use her sigla without brackets except when I cite her words directly. Page references to the dissertation appear parenthetically.
48 A little more than half of Bn exists in MS Berlin, Preuss. SB, 130; MS Paris, BNF, n.a.l. 1615; and MS Vatican, Reg. lat. 644.
49 Gorman, “Glosses”, p.210.
50 Lipp presented the texts of T on DNR 3 and 23 and of Bn and Bt on DNR 12 and 20 on p.240-259 of her dissertation.
51 Gorman, “Glosses”, p.223.
52 In a postscript, ibid., p.232, Gorman saw the possibility of using the new edition of Byrhtferth’s computistical works (note 25 above) to test his defense of Herwagen, but did not pursue it. Byrhtferth’s editors noted that the DTR glosses in Byrhtferth’s computistical collection, glosses that may be Byrhtferth’s very own, “have little or nothing in common with the ‘Auxerre glosses’ formerly attributed to B”. See Byrhtferth’s Enchiridion (ed. Baker and Lapidge), p.lxxxviii and p.liii-lv. See also Wallis, Bede: The Reckoning of Time (note 8 above), p. xcv, note 270.
53 PL 90, c. 236B; c. 241B; c. 403D; c. 412C; 417A. See Byrhtferth’s Enchiridion (ed.Baker and Lapidge), p.lxxxvi.
54 PL90, c. 356D: “Antiqui autem Anglorum populi. Angli nomen est regionis, in qua primum coeperunt populi Angli esse. Lege historiam Ecclesiasticam gentis Anglorum; ibi enim, in libro I, cap. 15, qualiter praedicta gens, anno ab Incarnatione Domini 449, in Britanniam primo advenit, per ordinem invenies”. This, the only gloss on the entire chapter, sounds like the work of a foreigner, not an Anglo-Saxon. Compare the glosses of Martin Hiberniensis which begin with the opening sentence later compiled in B (“Angul nomen regionis in qua primo ceperunt populi esse”), but which continue for more than two pages in CCSL 123B, p.329-332. Of these glosses, Jones remarked, “Purely continental commentators (e.g. MS No.128) do not comment on DTR, c. xv (De mensibus Anglorum), but our glossator does, without betraying any special knowledge of the subject. This fact suggests interest in Insular culture” (ibid., p.261; the emphasis is Jones’s). Martin did seem to have some special knowledge of Anglo-Saxon and could interpret Thrilidus, Giuli, and Salmonath.
55 PL 90, c. 304D, c. 477B, c. 484B. See Jones, “Byrhtferth Glosses”, p.93, and Byrhtferth’s Enchiridion (ed.Baker and Lapidge), i. 3.14, 34, 57; ii.1.149, 174; ii.3.243; iii.2.134; iii.3.132; iv.1.327.
56 Byrhtferth’s Enchiridion (ed.Baker and Lapidge), p.367, note 11.
57 The B glosses on DTR, on the other hand, use subaudis and subauditur (PL 90, c. 302D, c. 310C, c. 318D, c.476D), favorite expressions in the pedagogical lexicon at Auxerre.
Auteur
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