romote,- and support: research, publication, and :
Ay
and other institations hot Restricted. to
2
cligile
fe:
SPECULUM
A JOURNAL OF MEDIAEVAL STUDIES
Volume V, Number 1]
January, 1930
Published Quarterly by THE MEDIAEVAL ACADEMY OF AMERICA CAMBRIDGE, MASSACHUSETTS
~ 2 Gar. art < 2 | ~ t+ ~~ bog pe ¢ a r ‘
EDITORIAL BOARD Editor-in-Chief JEREMIAH Denis Maruias Forp
Managing Editor Francis Peasopy Maaoun, Jr
Assistant Managing Editor SAMUEL Hazzarp Cross
Publishing Editor Joun Nicotas Brown
CHARLES Rurus Morey Princeton University
Dana CARLTON Munro Princeton University
JaMES Hueco Ryan Catholic University of America
JouHN StronG Perry TatTiock
University of California
Ernest Hatcu WILKINS Oberlin College
Kari YounG Yale University
ADVISORY BOARD
ScouyLer ALLEN University of Chicago
CuaARLES HENRY BEESON University of Chicago
GrorGE CorrMAN Boston University
CorNELIA CATLIN COULTER Mount Holyoke College
Apams Cram Boston, Massachusetts
Gorpon Hatt GEROULD Princeton University
EtTIENNE GILSON Universities of Paris and Toronto
GrorGE LIVINGSTONE HAMILTON Cornell University
CuaARLES Homer Haskins Harvard University
GrorceE La PIANA Harvard University Joun MattuHews MANLY University of Chicago
Nitze University of Chicago
ArtuuR KinGsLeEY PorTER Harvard University
Epwarp KENNARD RAND Harvard University
Frep Norris Rosinson Harvard University
JAMES WESTFALL THOMPSON University of Chicago
Lynn THORNDIKE Columbia University
JAMES Fretp WILLARD University of Colorado
BUSINESS BOARD
Amos McManon
New York University
JOHN MARSHALL Cambridge, Mass.
GrorGE ARTHUR PLIMPTON New York City
The Editors should be addressed in care of the Mediaeval Academy of America, Cambridge, Massachusetts (cable address, ‘Speculum, Cam- bridgemass’). The attention of contributors is called to the ‘Notes for Contributors’ printed each year at the end of the January issue.
Vor. V, No. 1. — Copyright, 1930, by the Mediaeval Academy of America. — Printep rm U.S. A.
Entered as matter, May 8, 1926, at the Post Office at Boston, Mass., under the Act of August 24, 1912.
CONTENTS
The Ancient Classics in the Mediaeval Libraries J. S. Beddie
Le Codex Q. v. I. 6-10 de la Bibliothéque Publique de O. Dobiache-Rojdestvensky
The Bellum Troianum of Joseph of Exeter . W. B. Sedgwick Mediaeval Academy Excavations at Cluny, V. K. J. Conant The Wyclif Manuscript in Florence... . . . I. H. Stein
Dramatic Ceremonies of the Feast of the Purification K. Young
A Portrait of Albertus Magnus . S. F. Damon Noch Einmal ‘Rerum Dominis Pietas Semper Amica’ J. Balogh
A. J. App, Lancelot in English Literature, his Réle and Character (R. S. Loomis); K. Asakawa, The Documents of Iriki Illustrative of the Development of Feudal Institutions in Japan (H. F. Mac- Nair); H. F. Dunbar, Symbolism in Medieval Thought and its Con- summation in the Divine Comedy (C. H. Grandgent); J. B. Fuller, Hilarti Versus et Ludi (K. Young); A. F. Gombos, A Honfoglalé Magyarok Itéliai Kalardozdsa, 898-904 (A. Steiner); N. E. Griffin, A. B. Myrick, The Filostrato of Boccaccio, a Trans- lation with Parallel Text (C. H. Grandgent); F. Hoddick, Das Miinstermaifelder Legendar (G. H. Gerould); F. P. Magoun, Jr, The Gests of King Alexander of Macedon (H. R. Patch); J. F. Mountford, P. K. Baillie Reynolds, One Hundred Post-Classical Unseens (K. P. Harrington); R. E. Parker, The Middle English Stanzaic Versions of the Life of Saint Anne (G. H. Gerould); P. Sabatier, Le Speculum Perfectionis ou Mémoires de Frére Léon, sur la seconde partie de la Vie de Saint Frangois d’ Assise (J. F. Wil- lard); W. P. Shepard, La Passion Provengale du Manuscrit Didot (E. L. Adams); P. Taylor, The Latinity of the Liber Historiae Francorum (H. M. Martin); T. F. Tout, Chapters in the Adminis- trative History of Mediaeval England: the Wardrobe, the Chamber, and the Small Seals (J. F. Willard); J. Vielliard, Le Latin des Diplémes et Chartes Privées de ' Epoque Merovingienne (H. M. Martin); C. C. J. Webb, Ioannis Saresberiensis Episcopi Car- notensis Metalogicon (S. H. Thompson); E. Wessén, De Nordiska Folkstammarna i Beowulf (K. Malone).
Announcement of Books Received Notes to Contributors
3
49 17 | 95
cr
102 103
A JOURNAL OF MEDIAEVAL STUDIES
THE ANCIENT CLASSICS IN THE MEDIAEVAL LIBRARIES
By JAMES STUART BEDDIE
S THE twelfth century was the age of the second revival of the ancient classics, it would be of interest to know what works of Greek and Roman writers were present in the libraries of the period and which of those available had the greatest popularity and the widest diffusion. For this purpose a retracing of the manu- script tradition of the texts of the classical authors through the mediaeval period would be most desirable, indeed necessary, for a complete picture. Still, considerable information about the range of the mediaeval mind and its interest in the Greek and Roman writers may be gained from an examination of the catalogues of the twelfth- century libraries.’
Most of the mediaeval catalogues, including more than a hundred from the twelfth century, have now been printed. A list of the known catalogues up to the year 1500 was compiled by Th. Gottlieb in his register, Ueber Mittelalterliche Bibliotheken (Leipzig, 1890). Sev- eral collections of catalogues have been published, notably that of Becker, in which 136 lists of the twelfth century and earlier are re- printed.? A number of catalogues of French libraries are in L. De- lisle’s Cabinet des Manuscrits de la Bibliothéque Nationale.’ Collections of the mediaeval catalogues of the German and Austrian libraries are now in course of publication under the auspices of the academies of Vienna, Berlin, Géttingen, Leipzig, and Munich, those of the Aus-
1 On the thirteenth century, E. K. Rand, ‘The Classics in the Thirteenth Century,’
tv (1929), 262. 2 G. Becker, Catalogi Bibliothecarum Antiqui (Bonn, 1885). Paris, 1868-1881.
~
4 Ancient Classics in the Mediaeval Libraries
trian libraries under the editorship of Theodor Gottlieb and his suc- cessors, and those of the German libraries by Paul Lehmann.!
Even these formal inventories, however, are often unsatisfactory and have important limitations for use as sources for the number and contents of the libraries. Indeed, a number of libraries of the highest importance, such as those of Tours or St Albans, have left no catalogues or their catalogues have been lost. Nor have we a catalogue of the papal library before 1295.
Often these catalogues are mere check-lists or inventories of the library’s possessions, entered upon the fly-leaves of manuscripts or on other spaces empty and available. Account was usually taken only of the number of manuscripts, rather than of the works con- tained therein; and volumes were regularly listed by the name of the first work contained, in cases where several works are found in one manuscript those after the first being allowed to go uncatalogued.* The volumes are often found cited by brief titles, which are some- times indefinite, as Liber Virgili or Diuersorum auctorum liber unus.
Also, many of these lists have come to us with insufficient indi- cations of provenance and can be identified only as within the life- time of an abbot or bishop, or within fifty or a hundred years by palaeographical means.
Though the catalogues are thus often inadequate and indefinite, and though their testimony regarding the popularity of individual authors may sometimes be misleading unless taken in connection with the history of the manuscript tradition of their works, yet a perusal of the catalogues may be made to yield much of interest regarding the libraries, the authors therein, and their interest to mediaeval men.
The greater part of the content of the monastery and cathedral libraries was composed of ecclesiastical books: volumes of the Bible, scriptural commentaries, works of the Fathers, and books for divine service. In a number of cases the library was entirely theological
1 Mittelalterliche Bibliothekskataloge Osterreichs, vol. 1 (Vienna, 1915); Mittelalterliche Bibliothekskataloge Deutschlands und der Schweiz, vols 1, 11 (Munich, 1918, 1928).
2 A recent writer suggests that to arrive at the actual number of works in the library one
must multiply the number found in the catalogue by about four. D. H.S. Cranage, The Home of the Monk (Cambridge, England, 1926), p. 5.
Ancient Classics in the Mediaeval Libraries 5
and liturgical, and in the greater part of the libraries the non-ecclesi- astical content did not reach one third of the total.
Most twelfth-century libraries contained something of the ancient classics, though their amount and proportion differed widely in the individual libraries. Appearances of Greek books in the western libraries are rare. The monks of the Greek monasteries of South Italy seem to have iived to themselves, and such works of the Greek classical authors as they posessed had no influence beyond the im- mediate region. Though the library of St Nicholas at Casole was famed throughout the Terra d’Otranto and contained a manuscript of Aristophanes, it was in contact with the East rather than the West, and this knowledge of Greek drama did not spread.’
Plato was known indirectly, in the main. Of his works only the first part of the Timaeus in the translation of Chalcidius was avail- able at the opening of the twelfth century. Notices of this in the library catalogues are fairly numerous. It was present at St Gall, Lorsch, Hamersleven, Bamberg, Rastede, Tegernsee, St Bertin, Bec, Anchin, St Amand, Corbie, Engelberg, Salzburg, Whitby, Durham, Reading, Canterbury, Halberstadt, and elsewhere. Notices of the translation of the Meno and Phaedo made by Aristippus of Catania about 1156 do not occur before the date of the Biblionomia of Richard of Fournival, which describes the Phaedo.? The twelfth-century cata- logue of Anchin cites Plato de cosmopio, possibly a description of the Timaeus.
Aristotle had been more popular in the early Middle Ages because, thanks to Boethius, his works chanced to be better preserved. Of
1 The library of Casole is described by C. Diehl in Mélanges d’ Archéologie et d’ Histoire, v1 (1886), 173, and by K. Lake in Journal of Theological Studies, v (1903-1904), 33.
2 The Biblionomia, Richard of Fournival’s model catalogue of a model library, is printed by Delisle, Cabinet des Manuscrits, 11, 518-533. Richard was chancellor of the church of Amiens, his life falling within the period 1201-1260. His work purports to be, as explained in allegorical fashion in the introduction, a plan of education for the youth of Amiens, and espe- cially a plan for the formation of a library. He lists 162 books, with remarkably full descrip- tions, and gives a scheme of classification to aid the librarian in finding them readily. The studies of Delisle (Cabinet, 111, 387) and of Aleksander Birkenmajer, Bibljoteka Ryszarda de Fournival i jej Pozniejsze Losy (The Library of Richard de Fournival and its Ultimate Fate), Cracow, 1922, have identified a number of items in the Biblionomia with manuscripts described
in the mediaeval catalogues of the Sorbonne library and bequeathed to the Sorbonne by the theologian, Gérard d’ Abbeville.
6 Ancient Classics in the Mediaeval Libraries
the logical works available at the opening of the twelfth century the Categories were spread most widely, references to this treatise being found in the catalogues of Bobbio, Montier-en-Der, St Emmeram’s, Hamersleven, Pfiiffers, Reichenau, Wessobrunn, St Amand, Anchin, Arras, Reisbach, and, with the rest of the Organon, at Canterbury and Rochester. The De Interpretatione was listed at St Emmeram’s, Pfaffers, Salzburg, Anchin, Arras, and Fleury. The other works of the Organon, which made their appearance in the second quarter of the twelfth century, found their way into the library inventories slowly. The Prior and Posterior Analytics are listed in the catalogue of Engelberg by 1175 (this monetary possessed a very complete col- lection of logical works), at Glastonbury, and in the Biblionomia. In the twelfth-century catalogue of the library of the monastery of St Peter’s at Salzburg appears the item: ‘Metaphysica et topica Aristotelis.’' In a list of textbooks probably in use at Paris near the close of the twelfth century mention is made of the logical trea- tises of the Organon, the Metaphysics, the De Generatione et Corrup- tione, and the De Anima.’ In general, however, the notices of the scientific works of Aristotle in the catalogues are slow to appear and are not numerous. It is probable that their spread received a tempo- rary setback when, in 1210, a provincial council forbade the public or private reading of the works of Aristotle at Paris, this prohibition being reénacted and made to include the Metaphysics by the statutes of the papal legate in 1215; nor was it lifted before 1231, when Pope Gregory IX forbade their use until they had been purged of error.
Though the works of several of the classical authors were not looked upon by the church with complete approval, this temporary check to the spread of the works of Aristotle is the nearest approach in the Middle Ages to a prohibition of the use of any Greek or Roman author. The logical works of Aristotle and Porphyry, ac- companied by the commentaries of Boethius, appeared in numbers in the monastery libraries, but the scientific and metaphysical works, with the commentary of Averroés, did not make their way so easily.
1 G. Becker, op. cit., p. 234. 2 C. H. Haskins, Studies in the History of Mediaeval Science (2d ed., Cambridge, 1927), p. 373.
| I
Ancient Classics in the Mediaeval Libraries 7
A good many items appearing in the catalogues as the works of Boethius undoubtedly represent works of Aristotle in his translation.
The Greek medical works were known chiefly in the versions of Constantine the African, the monk of Monte Cassino, who translated the Aphorisms of Hippocrates and the Tegni of Galen, and of Bur- gundio of Pisa, who translated ten treatises of Galen.' The works of Hippocrates appeared in the library catalogues of S. Angelo, near Capua, St Amand, Durham, the medical library of Bishop Bruno of Hildesheim,? and elsewhere. Writings of Galen are mentioned in the catalogues of Reichenau, St Amand, Durham, Salzburg, Hildesheim, and others. The work of Dioscorides on herbs appears in the cata- logues of St Amand, Durham, and Peterborough.
Philo Judaeus is mentioned in the catalogues of St Riquier and Lyre, and in the course of the twelfth century a letter to Abbot Conrad of Tegernsee requested the loan of a manuscript of this work.
Josephus’ History of the Jews was regarded in the Middle Ages as a sort of auxiliary to the study of the Bible, and manuscripts of it in translation were diffused very widely, the work being recorded more than forty times in the catalogues.
The mention of two books of Lucian in the twelfth-century cata- logue of St Bertin is obviously an error in writing Lucan.’ The Bucolics of Theocritus are listed in the inventory of Pfiiffers made in 1155, but it is probably to the Eclogues of Nemesianus that this description refers.‘
In all these works of Greek authors available to the Middle Ages through translations, there was nothing, it may be observed here, which could give a correct conception of Greek life, which, along with the Greek language, remained largely an unknown field to the twelfth century.
The works of the Latin authors were much more widely spread, and most libraries possessed a larger or smaller selection from the Roman writers. Though the body of material then available was
1 Haskins, op. cit., p. 208.
2 K. Sudhoff in Archiv fiir die Geschichte der Medizin, 1x (1916), 348.
3 G. Becker, op. cit., p. 183.
* Lehmann, op. cit., 1, 486; Manitius, ‘Philologisches aus alten Bibliothekskatalogen,’ Rheinisches Museum, N. ¥., XLvu1, Ergiinzungsheft (1892), 57.
|
|
8 Ancient Classics in the Mediaeval Libraries
much the same as that possessed to-day, an inspection of the library catalogues will show that the preferences of the Middle Ages for individual authors differed in many respects both from those of the Romans and from those of the present day.’
The writings of Plautus (the Aulularia) are cited in the catalogues of Passau in the tenth century, Metz in the eleventh, and in that of the cathedral library of Bamberg in the thirteenth, but the reference is probably to the later Aulularia with the ascription to Plautus added. The catalogue of Michelsberg lists ‘Plauti liber I,’ which may refer to the same. The twelfth-century catalogue of Lambach names the Amphitruo.
The comedies of Terence were better known, but imperfectly understood. Mediaeval information about the ancient drama was drawn from the works of Isidore of Seville and Donatus. The inepti- tudes and uncertainties of the commentaries of Eugraphius and his successors show that a true conception of Roman comedy had been lost.2, Terence’s metre was especially puzzling, and his work was often treated as prose. Alberic of Monte Cassino cites Terence as an authority on style. Nearly twenty citations of Terence occur in the French catalogues before 1250, including three copies at Cluny. Few of these name individual plays, but Richard of Fournival has com- plete information, stating that the comedies number six, of which he supplies the names. A still larger number of manuscripts is noted in the catalogues of German libraries. That the plays were read is shown by the fact that the manuscript of Terence at Freising in the eleventh century was noted as being out on loan, as was that
1 Statistics relating to the diffusion of the works of the individual Roman authors in the mediaeval libraries have been assembled by M. Manitius in ‘Beitriige zur Geschichte der Rémischen Prosaiker im Mittelalter,’ and ‘Beitrige zur Geschichte der Rémischen Dichter im Mittelalter,’ in Philologus, vols xtvii—Lv1 (1888-1897), and Supplement vir (1899), also in ‘Philologisches aus alten Bibliothekskatalogen,’ Rheinisches Museum, xiv (1892), Erginzungsheft, 1-152.
Citations of classical authors in English library catalogues are listed by E. A. Savage, Old English Libraries (London, 1911), p. 258.
2 E. K. Rand, ‘Early Mediaeval Commentaries on Terence,’ Classical Philology, rv (1909), 359; Wilhelm Cloétta, Komédie und Tragidie im Mittelalter (Halle, 1890), p. 14; M. Schanz- C. Hosius, Geschichte der Riémischen Literatur (4th ed., Munich, 1927), p. 121; J. D. Craig, Jovialis and the Calliopian Text of Terence (St Andrews University Publications, vol. xx11, St Andrews, 1927).
Ancient Classics in the Mediaeval Libraries 9
at Cologne, and to an abbess. In England, Terence appeared in the catalogues of Peterborough, Durham, Rochester, and Canterbury, and in Italy at Bobbio, Monte Cassino, and Treviso.
The works of Cicero, especially the rhetorical and philosophical works, were found in numbers of the catalogues, along with certain other works, such as the Rhetorica ad Herennium, which went under the name of Cicero and was cited in the catalogues of ten collections in France and five in Germany, and of which Cluny possessed three manuscripts. It seems likely that all of his works were not collected or even known at any one time or place. It was the ambition of Abbot Wibald of Corvey to collect the works of Cicero in one volume, and he did succeed in assembling the largest group of Cicero’s writ- ings of the twelfth century. John of Salisbury bequeathed his manu- scripts of the De Officiis and De Oratore to Chartres. Mention is made of the writings of Cicero in more than thirty catalogues of libraries from France and in a like number of German catalogues. The selections represented ranged from the nineteen manuscripts of Cluny, the sixteen listed in the Biblionomia of Richard of Fournival, and nine of the library of Bec, to the single item of the catalogue of Benediktbeuern, listing the De Senectute and De Amicitia. These last two, as school books, were cited very often. Certain of the orations occur more frequently than others, among them the Pro Rege Deiotaro. Manuscripts of the letters were comparatively rare, though three are listed in the twelfth-century catalogue of Cluny and they were also to be found at Lorsch. The Topica and the De Offictis occur frequently, while other works, such as the De Legibus (listed at Bec and Durham), are mentioned only rarely. Several of the catalogue descriptions do not permit identification of the par- ticular work described, while certain others seem to show an imper- fect knowledge of the author. Thus the cataloguer of St Gildas writes ‘Librum Tulli Cesaris de oratore.’ Outside France and Ger- many citations of Cicero in the library catalogues are less frequent. Of the Italian catalogues those of Bobbio and Monte Cassino list manuscripts of Cicero. In England works of Cicero were listed in the catalogues of Durham, Whitby, Rochester, and Evesham. The catalogue of Canterbury (ca. 1170), containing references to the
|
:
|
10 Ancient Classics in the Mediaeval Libraries
De Amicitia, De Senectute, Rhetorica, and Topica, and that of Glaston- bury (1247), in which the De Senectute and De Amicitia occur, may be added to Manitius’ list of English citations. Apparently no twelfth-century Spanish catalogues mention works of Cicero.
The Verona manuscript of Catullus does not appear between the tenth and the fourteenth centuries.
Caesar is mentioned infrequently in the catalogues of mediaeval libraries. Of the fifteen or so naming his works, twelve are catalogues of French libraries. Manuscripts were also listed at Wiirzburg and Ripoll. No English catalogue includes Caesar before 1443.
Lucretius evidently had little appeal for the mediaeval mind, as surviving manuscripts are few and his work is cited only in the cata- logues of Murbach, Bobbio, and Corbie.
Catalogue mention of the histories of Sallust was frequent, his name being found in the catalogues of almost twenty French libraries, and in more than that number in Germany, in England at Canter- bury, Durham, Rochester, and Glastonbury, and in Spain at Silos.
These were what the twelfth century had from republican Rome, and their relatively small proportion in the catalogues is but another piece of evidence that to the Middle Ages Rome meant the Rome of the Empire.
The popularity of Virgil was universal. As a school author his work appears in every list of such books. The Eclogues and Georgics are named in the catalogues almost as often as the Aeneid, but the larger number of the citations of Virgil without naming the particular work probably refer to the latter. Thirty or more citations of the commentary of Servius attest the diligence with which Virgil’s works were studied. In the catalogues of the English libraries mention of Virgil is made in those of Bury, Durham, Lincoln, Canterbury; the Aeneid is named in those of Rochester and Glastonbury; the Bucolics in that of Whitby; and the Bucolics and Georgics in those of Durham, Reading, Rochester, and Glastonbury.
Mention of Propertius and Tibullus occurs only in the Biblionomia of Richard of Fournival, besides a notice of the latter in an uniden- tified French catalogue of the ninth century.
Horace is mentioned more than twenty times in French library
x
Ancient Classics in the Mediaeval Libraries 11
catalogues previous to 1250, and even more frequently in those of Germany. Cluny possessed two manuscripts of the complete works. The Biblionomia gives the names of the individual works. In Spain the works of Horace appear in the catalogue of Silos, and in Italy at Monte Cassino and Treviso. English catalogues which cite Horace are those of Bury, Canterbury, Durham, Rochester, Reading, and Glastonbury. Although the individual works are not often cited since the complete works could conveniently be contained in one volume, the Satires and Epistles appear slightly more frequently than the Odes.
Ovid, morally interpreted or otherwise, was even more popular. Nearly twenty French catalogues and almost thirty from Germany cite his works. In England they appear at Durham, Canterbury, Rochester, and Glastonbury, in Spain at Oviedo, and in Italy at Bobbio and Monte Cassino. These citations include a number of the pseudo-Ovidian works, such as the De Sompno.'- Durham pos- sessed no less than ten manuscripts of Ovid, Blaubeuern five, and Cluny three. The Metamorphoses was the most popular work, as its description as ‘Ovidius magnus’ or ‘Ovidius maior’ indicates.
Several of the poets of the Empire were held in greater esteem during the twelfth century than at the present day. Persius and Lucan served as school authors, the former being in addition regarded as a moralist and the latter as an historian. Persius appears in fifteen or more catalogues of French libraries and in more than twenty from Germany, at Canterbury, Whitby, Durham, Rochester, and Glastonbury in England, and at Bobbio in Italy. The works of Lucan appear in practically the same numbers. Four manuscripts are found at Corbie and at Bobbio, and two are listed at Cluny and in several other libraries.
The name of Petronius appears but once, and that in the cata- logue of an unidentified French library of the eleventh century. The frequency with which Martial was quoted would lead one to expect that manuscripts of his works would be more often listed than is actually the case, for they appear only about a dozen times,
1 P. Lehmann, Pseudo-antike Literatur des Mittelalters (Studien der Bibliothek Warburg, vol. xx111, Leipzig, 1927), pp. 89-91.
| 3 } | x 4 |
12 Ancient Classics in the Mediaeval Libraries
in France at Corbie and in the Biblionomia, in Germany at Lorsch, Bamberg, Muri, and Wiirzburg, in England at Peterborough, and in Italy at Bobbio.
Citations of Juvenal come most often from France, where he is named in twenty or more catalogues. About fifteen mentions of Juvenal occur in the German catalogues, seven in England, including Canterbury, Bury, and Glastonbury, while he is also listed at Bobbio, Oviedo, and Ripoll. Bobbio, St Bertin, and Rouen possessed three copies of Juvenal in their libraries.
Statius was widely read, quoted, and copied, that is, the Achilleis and the Thebais; the Siluae do not appear in the catalogues of the early Middle Ages. Twenty catalogues of the German libraries cite Statius, more than fifteen in France, five in England, and one (Silos) in Spain.
References to the Roman historians in the library catalogues illustrate the mediaeval preference for epitomes and condensations. Livy occurs in the catalogues of Limoges; in Corbie, which lists two copies of the third decade; in Cluny, which had two copies of the third decade and one of the first; in the collections of Murbach; in the cathedral library of Bamberg; and at Pomposa. Curtius Rufus’ name does not appear except in the catalogue of the library at Canterbury made after 1285, but notices of histories of Alexander the Great are numerous, whether reference is to his work or to those works associated with the names of Julius Valerius and the Archpresbyter Leo. Justin’s epitome of Pompeius Trogus is cited in half a dozen catalogues from Germany and in a like number from France, that of St Martial’s at Limoges listing two copies, and it is also named in the catalogues of Durham and Pomposa. Valerius Maximus is listed six times in France, but elsewhere only at Michelsberg. The epitome of Florus is catalogued at Chartres, Bec, Corbie, and Limoges, but in Germany only in the ninth-century inventory of Lorsch. No certain references to Tacitus are known, there being only the mention of a Histortam Cornelii cum Homero from Monte Cassino in the eleventh century. The Corbie catalogue of the end of the twelfth century has a Liber Cornelii de bello Troiano. Otto of Freising cites ‘Cornelius’ in his list of writers on Greek history.
( 1 t C 0 Pp Se H in Px ou Se TI as St of: Ages (Oxf
Ancient Classics in the Mediaeval Libraries 13
The works of Suetonius were known in France from the time of Lupus Servatus or before.! They are found listed in five catalogues from France, Bec possessing two manuscripts, at Bamberg and another (unidentified) library of Germany, and at Bury, Rochester, and Glastonbury in England. Though John of Salisbury made use of Suetonius in the Policraticus, he could on one occasion make two persons of Suetonius and Tranquillus.? The collection of biographies known as the Augustan History was listed in the catalogues of Mur- bach and the cathedral library of Bamberg, while Cluny possessed a volume containing the lives of the Caesars from Augustus to Theo- dosius. The history of Eutropius was cited in the catalogues of Pomposa, Durham, St Emmeram’s, Bamberg, and in France at Cluny, St Amand, Chartres, and Bec, which possessed two copies. Thus it will be seen that the work of no one of the classical historians was found among the more popular books of the twelfth century and none of them approached being generally known in the sense that the classical poets were known. The catalogue citations of the Seven Books against the Pagans (Historia aduersus Paganos) of Orosius, the first great Christian historian in the West, surpass those of any of the pagan writers on the subject.
The total of the citations of the works of Seneca the Younger is probably increased somewhat by inclusion of writings of the elder Seneca, and pseudo-Senecan writings, of which there was a number. He is mentioned in the catalogues of fifteen or more French libraries, in a somewhat smaller number from Germany, at Monte Cassino, Pomposa, and Citta di Castello in Italy, and at Durham, Peterbor- ough, Bury, Reading, and Glastonbury in England. Bec had seven Seneca manuscripts and the Biblionomia has a full list of his writings. The work most commonly cited was the Moral Epistles, often listed as ‘Seneca maior.’ The Apocolocyntosis appears in the catalogue of St Amand and in the Biblionomia. The apocryphal correspondence of Seneca and Paul is also listed frequently.
1 E. K. Rand, ‘On the History of the De Vita Caesarum of Suetonius in the Early Middle Ages,’ Harvard Studies in Classical Philology, xxxv11 (1926), 20. 2 Policraticus, v111, 18, and the note of C. C. J. Webb in his edition of the Policraticus
(Oxford, 1909), 11, 364.
|
4 4
|
Ancient Classics in the Mediaeval Libraries
14 Quintilian, because of the advanced character of his work and its attention to oratory, was not the rhetorical text of the twelfth century, though he seems to have been well known to John of Salis- bury. Five French catalogues cite the work, in addition to those of Durham, Bamberg, and Salzburg.
Citations of the Letters of Pliny the Younger are quite rare, and occur in the catalogues of Lorsch, an unidentified German library catalogue of the eleventh century, and in those of Angers and Bec in the twelfth century. Some confusion may have occurred with Pliny the Elder, and some of the items mentioning Pliny without further description may refer to manuscripts of the Letters.
The author of the Attic Nights was regularly described in the catalogues as ‘Agellius,’ and appears thus in the inventories of Cluny, Egmond, and Bamberg, in the Biblionomia, and in the corre- spondence of Reinald of Hildesheim with Wibald of Corvey regarding a loan of the volume.
Apuleius was known to the twelfth century as the author of the Periermeniae, cited in nine catalogues from France, four from Ger- many, and at Bobbio in Italy, where the catalogue lists two manu- scripts of Apuleius. The catalogue of Bec lists the De Deo Socratis and the De Fato.
The works of the Roman writers on the sciences and useful arts continued to be copied and read during the Middle Ages. The pref- erence here too, however, was for epitomes and compends. The work of Hyginus on astronomy appears in five French library catalogues, in nine from Germany, of which all but two are from the eleventh century or before, also in the catalogues of Durham, Rochester, and Treviso. Vitruvius on architecture occurs five times in catalogues from Germany, and in four French lists. Citations of Columella are confined to France, the work being listed in the catalogue of Corbie and in the Biblionomia. The same is true in the case of Pomponius Mela, cited only at Bec and Limoges. For so extensive a work the Natural History of Pliny is noted surprisingly often, being mentioned in eight catalogues of French libraries, in the same number or more from Germany, and in those of Bobbio and Pomposa in Italy. In several cases, however, it is stated that the library possessed less
(
i
]
Ss
e
a:
tk
Ancient Classics in the Mediaeval Iibraries 15
than the complete work. Solinus, who used Pliny’s work, is men- tioned with even greater frequency. The work of Serenus on medi- cine appears in the catalogues of St Riquier, Cluny, Rochester, Reichenau, Murbach, Weihenstephan, and Géttweig. Palladius’ work on agriculture was extremely popular in France and the Nether- lands, thirteen catalogues from this region listing the book. It also appears in the catalogue of Durham. Vegetius’ De Re Militari is listed in nine French library catalogues and in six from Germany. The Mulomedicina was listed at Reichenau.
The old Roman textbooks on grammar continued to be used throughout the Middle Ages, and catalogue citations of the works of Victorinus, Donatus, Martianus Capella, Priscian, and Phocas are numberless. Besides their use as texts they were of value in preserv- ing extracts and quotations from the works of the classical authors. The Distichs of Cato, another school text, was widely diffused. Eleven citations of it occur in catalogues of French libraries, more than twice that number from Germany, while it is also found in the libraries of Oviedo, Bobbio, Durham, Rochester, and Glastonbury. The Fables of Avianus appear in similar proportions. The Latin Jliad or ‘Homer’ is named in the catalogues of half-a-dozen libraries in France, and more than that number in Germany, also at Monte Cassino, Durham, and Whitby. Dares’ History of the Trojan War
_ was almost as widely circulated. The History of Apollonius of Tyre
is mentioned in the inventories of Rochester, Canterbury, St Wan- drille, Lobbes, Gorze, Cluny, St Amand, Pontigny, Toul, Stavelot, Reichenau, Weihenstephan, Wessobrunn, and Salzburg.
Certain of the authors of the last days of the Empire attained popularity in the mediaeval libraries. The Letters of Symmachus were listed in the catalogues of St Bertin, Angers, Bamberg, Arn- stein, and Hamersleven. The compendious works of Macrobius were extremely popular. His name occurs more than twenty times in the catalogues of French libraries and in almost as many from Germany, as well as in those of Durham, Rochester, Bobbio, and Ripoll. The commentary on the Somnium Scipionis is cited somewhat oftener than the Saturnalia.
Though the twelfth century marked the height of mediaeval
l |
| 4 |
| |
16 Ancient Classics in the Mediaeval Libraries
interest in the classics, they were by no means displaced in the fol- lowing age. We come upon catalogues of libraries of the thirteenth century, such as that of Pontigny, which had an extensive selection from the works of the Roman writers, including Caesar, Seneca, Quintilian, Palladius, Valerius, Maximus, Solinus, and Suetonius.! Conclusions as to the prevalence and popularity of the classics in the libraries of the thirteenth century must, however, be guarded, since our catalogues of the period after 1250, especially from the monastic houses are much less numerous than from the preceding period.
In the thirteenth century, however, the classics were forced to divide the attention of scholars with newer contenders, with logic and metaphysics, and with the new science, whose followers were indifferent to literary form and which introduced a technical vocabu- lary of its own.
The catalogues illustrate these new interests and tendencies and others as well. The thirteenth-century catalogue of the Cistercian house of Pontigny not only testifies to a continued interest in the classics, but it shows as well the loss of intensity of feeling in the Cistercian order, for in no Cistercian catalogue of the twelfth century can we find an extensive list of books dealing with non-religious subjects.
The catalogues indicate the library centres where interest in the classics prevailed especially. Thus Cluny, rich in everything, pos- sessed, according to the catalogue of the middle of the twelfth cen- tury, a large number of works of the classical writers among its 570 volumes.
The catalogues preserve to us the memory of mediaeval classical students, some famous, some forgotten. Thus the lists of the books in the hands of the monks of Durham? indicate the reading of Guarinus, a student of the classics otherwise unknown to fame, while the record of the bequest of the library of John of Salisbury to Chartres recalls the interests of the greatest classical scholar of the twelfth century.
1 Catalogue Général des Manuscrits des Bibliotnéques Publiques des Départements (Paris,
1849), 1, 697. 2 G. Becker, op. cit., p. 244.
| | | a 1 1 1 1
Ancient Classics in the Mediaeval Libraries 17
APPENDIX
In the following appendix is given a list of the library catalogues of the
period 1050-1250, which have appeared in print, but are not listed in Theodor Gottlieb, Ueber Mittelalterliche Bibliotheken, or which have ap- peared in later editions since the publication of Gottleib’s register (1890). It does not include, however, those printed in the series Mittelalterliche Bibli- othekskataloge Deutschlands und der Schweiz and Mittelalterliche Bibliotheks-
kataloge Osterreichs.
1
14
Altenburg 1200 Monastery (Benedictine). F. L. Helmling, Studien und Mitteilungen zur Geschichte des Benediktiner Ordens, x111 (1923-1924), 236. Altzelle xmthcent. § Monastery (Cistercian). Neues Archiv fiir Sdichsische Geschichte, xv111 (1897), 201. Aversa 1119 Grant by bishop to monastery of S. Lorenzo. Regit Neapolitani Archivi Monumenta (Naples, 1857), vol. v1, no. 574. Bamberg 1112-1147 Monastery of Michelsberg. Catalogue and lists of acquisi- tions (Gottlieb, nos. 12-14). H. Bresslau, ‘Bamberger Studien,’ Neues Archiv, xx1 (1896), 143-154. Bamberg x1th cent. Michelsberg (Gottlieb, no. 18. Becker, no. 80). H. Bresslau, art. cit., p. 165. Bamberg xuuth cent. Cathedral. List of books entrusted to Master Richard (Gottlieb, no. 20). H. Bresslau, art. cit., p. 170, note 11, p. 194. Barcelona 1142 Gift of Bishop Arnald to the Cathedral. R. Beer, Handschriftenschdtze Spaniens (Vienna, 1894), p. 66. Bari 1067 ~— Service books of Chiesa del Catapano. N. Tamassia, Atti del R. Istituto Veneto, ux1v, ii (1904-1905), 286 Beauvais xith or xitth cent. Gift of Roscellinus to Cathedral library (Gottlieb, no. 968). H. Omont, ‘Recherches sur la Bibliothéque de l’Eglise Cathedrale de Beauvais,’ Mémoires de I’ Académie des Inscriptions et Belles-lettres, xu (1916), 1. Beauvais 1217 ‘Gift of Bishop Philip to monastery library. Gallia Christiana, vol. 1x, col. 739. Bruges _—-xth cent. Monastery of St André. Revue Bénédictine, xxxvui (1925), 273. Burton after 1175 Monastery library (Gottlieb, no. 442). Zentralblatt fiir Bibliothekswesen, 1x (1892), 200 Bury ca. 1200 Catalogue of monastery of St Edmund (Gottlieb, no. 443). M. R. James, On the Abbey of St Edmund at Bury (Cambridge Antiquarian Society Octavo Publications, no. xxv111) (Cambridge, 1895), pp. 23-32. Bury xuithcent. Fragmentary list. G. F. Warner and J. P. Gilson, Catalogue of Western Manuscripts in the Old Royal and King’s Collections (London, 1921), 11, 158.
| | 4 | | 7 { 4 | | 4 | 8 ; 9 | 10 11 12 13 |
18 Ancient Classics in the Mediaeval Libraries
15 Canterbury ca. 1170 Christ Church (Gottlieb, no. 450). ‘ M.R. James, The Ancient Libraries of Canterbury and Dover (Cambridge, 1903), pp. 3-12. Facsimile. 16 Cava 1063 Church of S. Michel. 3 Codex Diplomaticus Cavensis (Milan and Naples, 1873-1893), vi11, 208. 3 17 Chartres 1150 Bequest of Thierry of Chartres to Cathedral. Chartulaire de Notre-Dame de Chartres (Chartres, 1865), 111, 206. 3 18 Citta di Castello 1143-1144 Gift of Pope Celestine II to the church (Gottlieb, no. 1369). A. Wilmart, Revue Bénédictine, xxxv (1923), 98. 19 Clairvaux xuth cent. Catalogue of monastery library (partial). 3 A. Wilmart, Mémoires de la Société Académique de l’ Aube, Lxxx1 (1917), 127. 20 Cologne ca. 1100 Church of St George. 4 Fr. Bock, Das Heilige Kiln (Leipzig, 1858), pp. 8-10. See KI. Loffler, ‘Kélnische Bibliotheksgeschichte,’ Zeitschrift des Deutschen Vereins fiir Buchwesen und Schrift- 4 tum, Iv (1921), 37. 21 Compostella 1226 Library of Archbishop Bernard. H. Omont, Bibliothéque de l' Ecole des Chartes, trv (1893), 327. 4 22 Durham 1095 Books of Bishop William de Karilef (Gottlieb, no. 1067). Republished with facsimile in H. D. Hughes, History of Durham Cathedral Library (Durham, 1925). Ebersberg after 1160 Monastery library. Der Katholik, uxxxvi11 (1908), 49. Flaxley xuth cent. Cistercian monastery (Gottlieb, no. 469). “
Zentralblatt fiir Bibliothekswesen, 1x (1892), 205. Gorze _xith cent. Monastery library. Revue Bénédictine, xx11 (1905), 1. Hildesheim ca. 1160 Medical library of Bishop Bruno.
K. Sudhoff, Archiv fiir die Geschichte der Medizin, 1x (1916), 348.
Lincoln 1150-1182 Catalogue of cathedral library and record of gifts.
Giraldus Cambrensis, ed. J. F. Dimock (Rolls Series, London, 1877), vu, 165-171; R. M. Woolley, Catalogue of the Manuscripts of Lincoln Cathedral Chapter Library ‘i 47 (London, 1927), p. v. &
28 Lincoln ca.1200 List of books and their donors to the cathedral library. R. M. Woolley, op. cit., p. ix, and again, with some variations, p. 144. Lugo 1230 Gift of Archdeacon Pedro to cathedral library (Gottlieb, no. 1346). R. Beer, Handschriftenschatze Spaniens (Vienna, 1894), p. 272. 80 Mainz 1186 Monastery of St Jakob. F. Schillman, ‘Wolfgang Trefler und die Bibliothek des Jakobskloster zu Mainz,’ Bethefte zum Zentralblatt fiir Bibliothekswesen, xi111 (1913), 24. 31 Marchiennes xith-xuth cent. Monastery of St Rictrude. ? 50 Analecta Bollandiana, xx1v (1905), 467. 3 82 Marienfeld ca. 1185 Cistercian monastery (Gottlieb, no. 119). # H. Degering, Beitrdge zum Bibliotheks- und Buchwesen Paul Schwenke gewidmet (Berlin, bs 51 1913), p. 53. Facsimile. a 89 Messina 1114 Bequest of Scholario Saba of the Basilian monastery of S. Salvatore di Bordonaro. Pirro, Sicilia Sacra (Palermo, 1733), 11, 1003. See F. Lo Parco, ‘Scholario-Saba,’ Atti della R. Accademia di Archeologia di Napoli, n.s., vol. 1, pt. 2 (1910), p. 207, and J. Heiberg, Byzantinische Zeitschrift, xx11 (1913), 160.
>
47
Ancient Classics in the Mediaeval Libraries 19
Messina 1173 Testament of the monk Clement of the Basilian monastery of S. Salvatore dell’ Acroterio. E. Aar, Archivio Storico Italiano, 4th Ser., rx (1882), 252. Namur 1218 Le Beffroi, 111 (1871), 129. Naples 1117 Treasure of church of St. Pietro. Regii Neapolitani Archivi Monumenta (Naples, 1857), v1, 35. Naples 1072 Church of S. Severo. Reg. Neap. Arch. Monumenta, vol. v, no. 416. Novalese xith cent. Monastery library. C. Cipolla, Memorie della R. Accademia della Scienze di Torino, 2d Ser., u (1901), 127. Novara 1175 Cathedral inventory. Revue des Bibliothéques, xx1 (1911), 106. Novara 1212 Two cathedral inventories. Historiae Patriae Monumenta, Chartarum, 1, 1192; 11, 1275. Ofia xuth cent. Monastery library. R. Beer, Handschriftenschdtze Spaniens (Vienna, 1894), p. 369. Paris xith cent. (close). Notre Dame (?) (Gottlieb, no. 422). Revue Bénédictine, xx1x (1912), 481, where Dom de Bruyne makes the ascription to Paris. Paris xurth cent. The Biblionomia of Richard of Fournival. L. Delisle, Le Cabinet des Manuscrits de la Bibliothéque Nationale (Paris, 1868-1881), 11, 513-535. Aleksander Birkenmajer, Bibljoteka Ryszarda de Fournival i jej Poznie- jsze Losy (The Library of Richard de Fournival and its Ultimate Fate) (Cracow, 1922). Perrecy xith cent. Two catalogues of a priory of Fleury. Revue des Bibliothéques, v1 (1896), 225. Poblet xuth cent. Carthusian monastery (Gottlieb, no. 743). R. Beer, Handschriftenschdtze Spaniens (Vienna, 1894), p. 405.
46 Pomposa 1093 (Gottlieb, no. 625, Becker, no. 70.)
G. Mercati, ‘I] Catalogo della Bibliotheca di Pomposa,’ Studi e Documenti di Storia e Diritto, xv11 (1896), 143. Reading 1208 Books returned to King John. Rotuli Litterarum Clausarum (London, 1833), 1, 108, Susser Archaeological Collections, 11 (1849), 134.
48 Reading xuth cent. Leominster Church, dependent on Reading Abbey.
51
English Historical Review, 111 (1888), 123.
49 Ripoll after 1046 (Gottlieb, no. 745.)
R. Beer, Handschriftenschdtze Spaniens (Vienna, 1894), p. 412, Vienna Academy, Sitzungsberichte, phil.-hist. Kl., vol. civ (1907-1908), no. 3, p. 101; vol. civimt (1908), no. 2.
Rochester ca. 1200
G. F. Warner and J. P. Gilson, Catalogue of Western Manuscripts in the Old Royal and
King’s Collections (London, 1921), 1, 308. Rolduc 1230 Monastery library.
Annales de la Société Historique de Maestricht, 1 (1854), 263.
S. Andreas de Rosans (Hautes Alpes?) xithcent. Books of Robert de Galone.
Zentralblatt fiir Bibliothekswesen, xx (1908), 370.
53 St Andrews 1144-1152 Gift of Bishop Robert to the prior and canons.
Haddan and Stubbs, Councils and Ecclesiastical Documents (Oxford, 1873), vol. 11, pt. 1, p. 227.
4 35 | : > 8 | | | 41 : a 42 | 44 | « 3 = q | |
20 Ancient Classics in the Mediaeval Libraries
54 §.Benigne xuthcent. Monastery library. Catalogue Général des Manuscrits des Bibliothéques Publiques: Départements (Paris, 1889), vol. v, p. iv. 55 §.Etienne cent. Monastery library. Notice in Bibliothéque de l’ Ecole des Chartes, tv (1894), 69. 56 S. Evroult cent. (Gottlieb, no. 396.) Ordéric Vital et ! Abbaye de Saint-Evroul, Notices et Travaux (Alengon, 1912). 57S. Martin de Stoppana 1156 Monastery inventory. P. de Bofarull y Mascaro, ed., Coleccién de Documentos Inéditos de la Corona de Aragon (Barcelona, 1849), rv, 241. 58 S. Nicolé di Trullas 1113 Gift to monastery. Historiae Patriae Monumenta — Codex Diplomaticus Sardiniae, 1 (1861), 89. 59 Salerno 1057, 1058 Churches of S. Felice and S. Nicola. Codex Diplomaticus C is (Naples and Milan, 1873-1893), vi11, 26, 66. Extracts from documents by N. Tamassia, Atti del R. Istituto Veneto, uxtv, ii (1904-1905), 273-286. 60 Selo xith cent. (close). Monastery of St Peter. Fr. Racki, ed., Monumenta Spectantia Historiam Slavorum Meridionalium. vu. Docu- menta Historiae Chroaticae (Agram, 1877), p. 181. 61 Silos cent. (Gottlieb, no. 749.) M. Férotin, Histoire de ’ Abbaye de Silos (Paris, 1897), p. 262. 62 Spalato xith cent. (close). Monastery of St Benedict. Fr. Racki, ed., Monumenta Spectantia Historiam Slavorum Meridionalium. vu. Docu- menta Historiae Chroaticae (Agram, 1877), p. 182. Tournel 1230 Gift of lords of Tournel to chaplain (Provengal). Revue des Sociétés Savantes, 6th Ser., v (1877), 205. Troja 1108-1187 _— Gifts of Bishop William to the cathedral (Gottlieb, no. 1229). L’ Arte, 3d Ser., rx (1906), 138. Val St Hugon 1224 Carthusian monastery. Bibliothéque de l' Ecole des Chartes, tv1 (1895), 662. Windberg xuth cent. (2d half). Premonstratensian canons (Gottlieb, no. 219).
Neues Archiv, xxx11 (1906-1907), 246.
In addition to the above there have been published several catalogues whose point of origin has not been determined.
& B
Germany xu1Ith cent. Zentralblatt fiir Bibliothekswesen, xx (1903), 370; Berliner Philol. Wochenschrift, xxx1x (1919), col. 984. France x11th cent. (close). C. H. Haskins, Studies in the History of Mediaeval Science (2d ed., Cambridge, 1927), p. 372.
England xuth cent. G. F. Warner and J. P. Gilson, Catalogue of Western Manuscripts in the Old Royal
and King’s Collection (London, 1921), 1, 133.
xith cent. (close). English Historical Review, xxx11 (1917), 388.
Uprer Iowa UNIVERSITY Fayette, Iowa.
|
=
4 x
i
4 3
4
2
5 i
LE CODEX Q.v.I. 6-10 DE LA BIBLIOTHEQUE PUBLIQUE DE LENINGRAD!
Par OLGA DOBIACHE-ROJDESTVENSKY
§ 1. La Description pu CopEex
ANS le fonds latin de la Petropolitana, actuellement Leninopo-
litana, il y a un MS. de haute ancienneté et de grande valeur qui, tout en gardant toujours sa place sur les rayons de la biblio- théque, indiqué d’une fagon exacte dans ses inventaires manu- scrits? et catalogues imprimés,’ a été, durant un siécle et demi, perdu de vue, oublié de la fagon la plus incompréhensible.
C’est le Q. v. I. 6-10, jadis N 840 de St-Germain, un MS. en onciale du vi‘ siécle qui contient un traité de Pseudo-Rufin, une épitre de Fulgence, deux homélies d’Origéne, et deux épitres de saint Jéréme. A partir de la fin du xvur* siécle, un silence se fait autour de ce MS.
Les Mauristes qui, en 1679-1700, ont publié les ceuvres de saint Augustin, en y insérant |’épitre en question de Fulgence (= Patr. Lat., xu, 752-778), ont collationné pour cette édition notre MS. ‘Corbeiensem uetustissimum.’ Pour les ceuvres de Pseudo-Rufin, de saint Jérdme, d’Origéne, ils ne ont pas apprécié (ci-dessous, p. 29). Dom René Tassin et Dom Ch.-Francois Toustain connaissaient ce MS. et l’ont décrit dans leur Nouveau Traité de Diplomatique (Paris, 1750-1765), 1, 154, 155, 171 suiv., 412, 416, 420 et Planches xu, v; XLIV, II; LVI, U, 2; Iv, 4; et v, 4. Une bréve notice de 18044 semble étre la derniére du x1x‘* siécle. Les ténébres se font autour du volume précieux. Personne ne s’en occupe. K. Gillert § dans sa description des MSS de St-Pétersbourg ne le mentionne pas, ni Traube—Leh- mann * non plus dans leur liste des MSS en onciale. Il est moins
1 Dédié a Ja mémoire vénérée de mon maitre et ami, Charles-Victor Langlois.
2 Voir L’Inventaire Manuscrit de la Bibl. Publ., Membr. Q. v. I. 6-10. 3 A. Staerk, Les Manuscrits latins du V° au XUI® siécle (St-Pétersbourg, 1910), 1, 6 sq.,
Pl. 1; u, Pls. v—-1x. 4 Allgemeine Litteraturzeitung (Halle, 1804), Intelligenzblatt, Nr. 77 (16 Mai 1804), p. 621,
§§ 5-9. 5 K. Gillert, ‘Lateinische Handschriften in St-Petersburg,’ Neues Archiv, v (1879), v1
(1881). 6 L. Traube, Vorlesungen und Abhandlungen (Munich, 1909), 1, 225-227.
21
| 3 2 i q | f ; | | | 4 i
22 Le Codex Q.v.I. 6-10 de Léningrad
étonnant (car leurs éditions ne prétendent pas épuiser la liste d’on- ciales) que ni chez Zangemeister—-Wattenbach,' ni chez Chatelain * il ne soit question de ce MS. Ce qui est plus singulier, c’est que Zim- mermann * n’en décrit ni n’en mentionne que trois volumes sur cing pour aboutir a des conclusions, 4 notre avis paradoxales, concernant les origines du codex. Hilberg,* qui a publié les épitres en question de saint Jéréme d’aprés des MSS dont le plus ancien est celui du rx* siécle, semble ignorer le Petropolitanus du vit (cf. ci-dessous, p. 29), et Lindsay dans ses Notae Latinae® en omet toute mention. Ecrit en onciale, le MS. n’entre pas dans ses recherches. Mais les marginalia auraient pu l’intéresser.
C’est aux érudits qui s’occupent plus spécialement des textes pré- sentés par notre MS. qu’il appartient d’apprécier son apport 4 leur connaissance. En attendant, il nous parait utile de communi- quer nos observations sur le MS. et nos conclusions a |’égard de ses origines. Nous débutons par une description sommaire. Entré au Musée Doubrovsky en 1792 (ci-dessous, p. 32), il fut ensuite partagé en cing volumes, pour lesquels Doubrovsky a commandé de belles reliures en maroquin rouge, bordé de vert et gravé de rinceaux d’or. Désormais chacune des ceuvres (Pseudo-Rufin, Fulgence, Origéne, saint Jéréme: Epistola ad Fabiolam, et saint Jéréme: Epistola ad De- metriadem) se présente comme un volume a part. Mais le MS., qui portait jadis 4 St-Germain les numéros 580 et 840 (ces numéros sont conservés sur le premier folio du premier volume), garde tou- jours, en dépit de la division, l’unité fonciére qui se révéle par nombre d’ indices paléographiques, artistiques, et littéraires.
Nous allons le décrire dans son intégrité.
Comme les beaux MSS du vi° siécle, c’est un volume de parche- min trés fin. Ses folios, au nombre de 220, mesurent actuellement 265 mm. X 205 mm. et le texte 170 mm. X 132 mm. II est done
1 K. Zangemeister-W. Wattenbach, Exempla Codicum Latinorum’ Litteris, Maiusculis Scriptorum, Heidelberg, 1876.
2 A. Ch&telain, Uncialis Scriptura Codicum Latinorum, Paris, 1906.
3 E. H. Zimmermann, Vorkarolingische Miniaturen, Berlin, 1916.
‘ I. Hilberg, Corpus Scriptor. Ecclesiast. Latinorum, tiv (1912), 49-87, et ivi (1918), 175- 201; abrév. CSEL.
5 W. M. Lindsay, Notae Latinae, Cambridge, Angleterre, 1915.
«
A i 3
4
» ‘MXP LINON D mene be
|
’ 7 | Se. | | | _
] SHONVIG
Ry) +)
‘a | 3 3 | t ~. 2 Oo? | \ . q r q oa 3
| -
Le Codex Q.v.I. 6-10 de Léningrad 23
presque carré:! encore une preuve d’ancienneté. Le nombre de lignes est toujours 22. Bien que le volume soit congu comme un tout, les groupes de cahiers destinés 4 des traités séparés s’y présentent in- dépendants, comme composition et numérotation. Les scribes de ces divers groupes travaillaient séparément: l’un n’avait pas épuisé son parchemin, que le suivant commengait un cahier nouveau.
Le scribe de Pseudo-Rufin, ayant 4 sa disposition 7 quaternions + 1 trinion, a laissé blanc le dernier folio (qui est coupé — a quelle époque?). Cela fait 61 folios, numérotés d’une facgon identique (1-61) en France et en Russie. Mais dans cette premiére partie, l’ancien scribe n’a inséré nulle part les indications des quaternions. Le vo- lume, comme les autres, est précédé d’un folio de papier, dont nous n’allons plus mentionner la présence.
Le copiste de Fulgence — 7 quaternions —a laissé blancs les quatre derniers folios, qui ont été ensuite coupés (au total ce sont 52 folios écrits, numérotés 62-113 en France, et 1-52 en Russie). Indépendant de son successeur 4 cet égard, il l’était également par rapport a son prédécesseur, car il a numéroté ses cahiers de quaternion 1 a quaternion v1, les derniers quatre folios n’étant pas numérotés.
Celui d’Origéne — 5 quaternions + 1 folio (numérotés 114-153 + 137°" en France, 1-41 en Russie) — les a numérotés d’une facon énigmatique: de quaternion XxvViII 4 quaternion XXXII.
Celui de saint Jéréme 1" — 4 quaternions + 1 trinion (numérotés 154-191 en France, 1-38 en Russie)—ne les a pas numérotés du tout.
Enfin, celui de saint Jéréme u*— 3 quaternions + 1 trinion, le dernier folio blane (coupé: au total 192-220 ff., numérotés ainsi en France, 1-29 en Russie) — les a numérotés de quaternion 1 4 quater- nion 111, en chiffres minuscules, élégants, ornés de légers traits.
L’encre, jaunatre, a pali, jusqu’au point d’étre parfois compléte- ment effacée. En dessus des traces de l’ancienne écriture, sur plu- sieurs folios (1, 24, 25, 40-41, etc.) le texte a été récrit (au xv‘ siécle?).
Un systéme uniforme domine |’emploi du rouge 4 travers le volume entier. On s’en sert avec beaucoup d’esprit pour mettre en relief le
1 Tl faut noter que la marge droite est coupée en partie.
1] q 4 | ‘a |
24 Le Codex Q.v.1. 6-10 de Léningrad
démembrement du texte: (1) pour les initiales ' de chaque alinéa (de quatre 4 sept, sur chaque page); (2) pour les trois-quatre lignes ini- tiales de chaque traité; (3) pour les mots importants au milieu du texte: deus, moyses, helia, saluator (ff. 101, 206, 207); (4) dans le traité de Fulgence pour Il’apostrophe, qui revient quarante fois: Firmissime hoc tene et nullatenus dubites; (5) pour les chiffres ro- mains en marge qui indiquent l’ordre (a) de ces exhortations, (b) des stations du peuple hébreu (ad Fabiolam); (6) enfin, pour les ex- plicits (ff. 61, 113, 131, 153, 191, 220), ot les lignes rouges alternent avec les noires; (7) on peut mentionner encore les guillemets rouges au début du traité Ad Fabiolam; et enfin (8) les numéros 1 et 1 au milieu du texte du f. 163.
Chaque traité débute par une initiale d’environ 70 X 30 mm. (IJ de 90 mm. X 3 mm.). Ce sont, au début de tous les traités, les ini- tiales H, E, O, I, A, I, I, en rouge, jaune, et vert-bleuatre, ornées d’entrelacs. Des petites fleurs aux pétales pointues (des lys?), des volutes trés fines, des ‘perles’ rouges minuscules, suspendues aux fils légers qui décorent les initiales en haut et en bas, font preuve d’un gott délicat et d’un pinceau fin, malgré les formes un peu lourdes des entrelacs qui remplissent les champs des lettres. I] faut noter que des J tout-i-fait analogues, avec entrelacs des mémes formes et des mémes couleurs, se retrouvent précisement dans les MSS italiens, comme I’attestent les planches et les descriptions de Zimmermann (voir op. cit., Planches 13 c, 15, 17, 19, 20, 22, et 32 et le texte pp. 146 suiv.). Les vignettes de notre manuscrit aux ff. 114, 153, 191, 220, sont composées de lignes brisées, jaunes et rouges, ornées de volutes et de pétales. Trés différent se présente le cadre autour de |’explicit au f. 113 avec ses entrelacs trés lourds, coloriés de vert trés vif, de jaune-brun et de rouge (cf. ci-dessous, p. 33 suiv. et la Planche tv).
L’écriture, bien qu’elle varie d’un traité a l’autre et ne soit pas le produit d’une seule et méme main, est pourtant d’une seule et méme école. Libre et élégante aux ff. 1-61 (Rufin), elle est molle, allongée, et inégale aux ff. 62-113 (Fulgence), pour revenir ensuite plutét au premier type et devenir gauche et négligée au cinquiéme traité. Malgré ces nuances, elle garde toujours le méme air de famille. Celui
1 Faites également en onciale: ce qui est propre aux manuscrits trés anciens.
|
zx
d
é
ic
fe
SC
el
pl
er
eu
n’
ho
Le Codex Q.v.I. 6-10 de Léningrad 25
qui confronterait nos Planches m et 11 (le premier folio beaucoup récrit) avec la Planche xv de Chatelain (Uncialis, cit. supra: col. I, Ambros. D 26 sup. Bobiensis. Prudence) serait saisi de l’air de pa- renté entre notre MS. et celui de Bobbio du vie siécle. Les formes identiques des lettres (A, F, L, T, P, R, S) dans les deux MSS, ainsi que leur aspect analogue, donnent le sentiment d’étre en présence de fréres jumeaux et se trouvent en accord avec les observations de Chatelain (Uncialis, p. 24) sur les MSS du vir siécle.
Aucune division des mots. On a essayé de la faire ensuite au moyen de petites virgules retournées (¢poque caroline: encre rous- sitre) et de traits (xvir* siécle? encre grise).
Un goat bien marqué pour les alinéas s’accuse, ainsi qu’une ten- dance 4 démembrer le texte colts et commatibus, a faire ressortir les citations en leur conservant des grandes marges, 4 disposer des séries de locutions bréves en colonnes:
quia ceruix tua quia oculi tui quia genae tuae, etc. f. 128 (cf. Planche u1.)
A
Une autre chose qui est digne d’étre notée, c’est le systéme achevé des distinctiones, observé fidélement:' des points toujours petits et élégants entre les phrases et leurs membres. L’encre de ces signes, identique a celle du texte, l'art avec lequel ils y sont insérés, pour faire avec celui-ci un ensemble organique, attestent que ces signes sont contemporains du MS. La ponctuation postérieure, celle de l’époque caroline (points et virgules, virgules retournées), faite en encre différente, et dérogeant 4 l’harmonie de l’écriture, révéle au premier abord sa nature hétérogéne.
La clarté de la disposition du texte, assurée par ces moyens, est encore rehaussée par l’emploi du rouge (ef. ci-dessus, p. 24).
Les explicits sont écrits dans notre MS. en capitalis rustica. I n’y a pas d’incipits proprement dits, si ce ne sont ceux de l’époque postéri- eure. L’hypothése qu’ils auraient disparu avec les feuilles coupées n’est pas exclue. Deux fois, au-dessus du texte, on reléve (dans les homélies d’Origéne et dans l’épitre de saint Jéréme, Ad Demetriadem)
1 Sauf, comme toujours, le dernier traité, surtout 4 partir du f. 208.
i
4
é
t
|
26 Le Codex Q.v.I. 6-10 de Léningrad
une espéce de titre en petites capitales. Les marginalia sont riches en onciales et cursives d’époques différentes.
En ce qui concerne les abréviations, tous les copistes sont égale- ment sobres. Ils ne connaissent que sept Nomina Sacra: ds, dns, ths, rps, sps, scs, isrl. Dns et scs n’ont pas nécessairement de sens sacré: hester loquebatur ad dnm (‘au roi’: f. 195 = CSEL, tv, 177); oli- barius ... clemens dns (f. 195 = CSEL, tv, 178); si spm potesta- tem habentes (f. 203 = CSEL, tv, 187). Mais spiritalis est écrit en toutes lettres et parfois aussi sanctus: sca sanctorum (f. 114), sancta trinitas (f. 63), une fois scificetur. Souvent, mais pas toujours, le que est remplacé par g. Mais le -bus est écrit en toutes lettres. Le m est remplacé par un trait la fin de la ligne. L’n ne l’est presque jamais.' Lettres conjointes (nt) sont rares. Tous les copistes écrivent l’o ex- clamatif avec un trait: 6 altitudo, 6 sponsa, 6 adulescentula (ff. 8, 129, 146).
Notre MS. a une particularité qui tient au contenu du texte et aux habitudes littéraires des auteurs latins grécisants et hébraisants, que les copistes suivent déja avec une peine évidente. Ce sont des locu- tions grecques et hébraiques, citées de temps en temps. II suffit qu’un siécle se passe pour que tout empire rapidement. Dans les copies du vir‘ siécle, les formes grecques dégénérent définitivement: les formes graphiques tout d’abord. Elles nous apparaissent décomposées, éclectiques: mi-grecques, mi-latines, jusqu’au jour ou la transcription latine s’y substitue pure et simple.
Cette dégénérescence apparait déja dans notre MS. Les efforts faits pour rendre les formes authentiques sont pénibles et imparfaits. Les mots grecs présentent un mélange curieux d’onciales et de capi- tales, de lettres grecques et latines. Ils ont des traits horizontaux au- dessus de chaque syllabe. Que l’on examine ces iléts grecs dans
etc. sont grecques, et les Z et a latines, les e et a sont onciales et
1 Le systéme est trés ancien: celui du vi® siécle au plus tard. Cf. L. Traube, Nomina Sacra (Munich, 1907), pp. 182-245; E. A. Lowe, ‘Some Facts about our Oldest Latin Manuscripts,’ et ‘More Facts,’ etc., The Classical Quarterly, xrx (1925) et xx11 (1928); W. M. Lindsay, Notae Latinae, les premiéres pages et passim.
a
h
q
;
pi
l’épitre Ad Demetriadem (= CSEL, v1): cvnupide (f. 199 = ed. p. 182); | d
alldpoc (f. 179 = ed. p. 75), KaKiac imac, ot lesw, p,c,
ke
qu
le
Le Codex Q.v.I. 6-10 de Léningrad 27
les M et N capitales; le mot spudasmatia est écrit en latin (f. 218 = ed. p. 200), et au mot eyv4 un agminis s’est substitué (f. 218).
Les procédés sont analogues dans la copie d’Origéne. Ce sont tou- jours les traits au-dessus de chaque syllable, la confusion des formes grecques et latines, onciales et capitales. Mais on remarque une ten- dance 4 éviter le grec. Au lieu du long passage de la Patr. Graeca, xu, 43, on ne trouve au f. 125 du MS. qu’un fragment mutilé: EIIICTHPHZOMENE, sed EIIICTIOHZOMENH. A la place du ‘owavbévor (Patr. Graeca, x11, 54) nous avons au f. 144 in uinaria. Le senti- ment latin commence a voiler les reminiscences grecques.
Pour les mots hébraiques on adopte le méme procédé des traits au- dessus de chaque syllabe. Ils sont écrits en lettres latines. Ces mots ne sont jamais déclinés; voir dans l’épitre Ad Fabiolam: hiroth, siph, défeca, halis, raphidim (ff. 159, 163 = CSEL, iv, 55, 59). Les traits horizontaux au-dessus de chaque syllabe ici, comme dans les mots grecs, ne font qu’indiquer les mots étrangers.'
En résumé: tout parle, dans notre MS., pour le vit siécle: la qualité du parchemin, les dimensions de la page, le caractére de l’écriture, la disposition du texte, le systéme des abréviations, et, plus particuliérement des m et n (cf. E. A. Lowe, op. cit.), les dessins et les couleurs des initiales.
On verra que l’écriture des gloses marginales et l’orthographe du codex plaide pour la méme époque (cf. pp. 30 suiv. et 40 suiv.).
§2. Le ContTENU
Nous passons au contenu du MS.
Ainsi que nous l’avons dit plus haut, il y a dans notre MS. cing parties, cing ceuvres différentes, qui sont actuellement comprises dans des reliures 4 part. Nous en allons décrire le contenu, une 4 une.”
1 L. Traube, Nomina Sacra, p. 48, tout en notant ce procédé dans les MSS en onciale du vit siécle par rapport aux mots hébraiques et grecs (‘in der Unciale war, wegen der Abnlich- keit der beiden Schriften, ein solches Verfahren fast unumgiinglich’), ne parait sousentendre que les traits au-dessus des mots entiers.
2? En transcrivant les textes, nous ne donnons des majuscules que si elles sont données par le MS. méme. Comme signes de ponctuation nous ne mettons que le minimum: quelques
| : | |
28 Le Codex Q.v.I. 6-10 de Léningrad
I ( Pseudo-Rufini de Fide)
Folios 1-61: Haec nostra fides est quam didici/mus a d{e}o et doctis uiris X similiter habentem cui/ gloria in saecula saeculorum. amen./ exp{licit] rufini pries)blytert] prouinciae palestinae/ liber de fide trans- latus de graeco in latinum sermonem. amen.
Le texte a été publié en 1650 par J. Sirmond,! d’aprés deux MSS corbéiens dont l’un est le nétre; en 1745 il a été reimprimé par Do- menico Vallarsi (= Patr. Lat., xx1, 1123-1154).
II (Fulgentii Ruspensis epistola de fide catholica )
Folios 62-113: epistulam fili petre tuae caritatis / accepi in qua te significasti te uelle hierosolymam pergere X fideliter perseueret et si quit/ aliter sapit et hoc quoque illi d[eujs reuelauit. amen./ fulgenti episcopt s{an]c[t]ae/ ecclesiae catholicae ruspensis de regula uerae fider libler] exp{licit).
Les Mauristes dans l’édition des ceuvres de saint Augustin, de 1679-1700 (= Patr. Lat., xu, 752-778) ont collationné notre MS. et en donnent les variantes.
III (Origenis in Canticum Canticorum homiliae duo beato Hieronymo inter prete ) (a) (Prologus beati Hieronymi ) Folios 114-114 v.: beatissimo papae damaso hieronimus./ Origenes cum in ceteris libris omnes uicerit X cum sic possint placere quae parua sint/. d{omi]no d{elo/ gratia sunt/ sae/ cu/. la/a/m/e/n.
(b) (homelia prima ) Folios 115-131: In exordium cantici canticorum X et eris uere diues in sponsi domo sponsa perfe/cta cui est gloria et imperium in sae/cula saeculorum amen/. cantici canticorum libfer] .I. exp{licit).
points, et ceci d’aprés notre systéme moderne. Le texte omis entre l’incipit et ]’explicit est symbolisé par une croix oblique. Les titres des traités, sauf dans les cas od ils rendent exacte- ment ceux du MS. original, sont pris entre parenthéses angulaires { ).
1 Rufini presbyteri prouinciae Palestinae liber de fide, etc., Paris, 1650.
| § I q I 2 le
Le Codex Q.v.I. 6-10 de Léningrad 29
(c) (homelia secunda )
Folios 131-153: Incipit etusdem liber secundus amen/. Ab eo loco in quo scribtum est/ nardus mea dedit odorem X ut digni efficiamur sponso serm/one sapientia xp{istlo th{es}u cut est imperium et gloria in saecula saecu/lorum amen/. expllicit] expositio origenis de/ cantico canticorum quod bea/tus hieronimus de graeco sermone in latinum transtulit. amen.
Notre MS. n’a été collationné ni par |’éditeur des ceuvres de Yauteur: Origéne (Ch. de la Rue, Opera omnia (Paris, 1733-1759) = Patr. Graeca, x11, 35-58), ni par celui des ceuvres du traducteur: saint Jéréme (Martianay—Pouget, 1593, revisé par Domenico Val- larsi = Patr. Lat., xxu, 1117-1143). Les divergences étant consi- dérables, une collation systématique s’impose.
IV (Beati Hieronymi epistola ad Fabiolam: De XLII manstonibus filiorum Israel per heremum )
Folios 153-191: In septuagensimo septimo psalmo que[m] turta euangelistam X praeutcatores autem corruent in illis. amen. qui legis/ ora/ pro/ m/e/. explicit] de mansionibus israhelitict populi. d[omi]no gratias.
L’édition du texte de Martianay—Vallarsi (= Patr. Lat., xx11, 698- 724) ignore notre MS. I. Hilberg, qui a, en 1912, publié ce texte d’aprés cing MSS (CSEL, tv, 49-87), dont le plus ancien est celui du Ix* siécle, ne le connait point non plus. Trés proche du texte de cette édition, notre MS. fournit pourtant des variantes intéressantes.
V_ beati hieronimi ad demetriadem/ uiriginem xplist}i inc{ipit]
Folios 191-220: Inter omnes materias quas ab adulescentia/ < quarum imitatio forma uirtutum est/ beati hiero/nimi ad demetriadem uirginem/ epistola exhortatoria/ exp{licit).
Notre MS. n’est collationné ni par Martianay—Vallarsi (= Patr. Lat., xx, 1107, No. exxx), ni par Hilberg, CSEL, tv1 (1918), 175- 201. Le texte, trés proche de celui de cette édition-ci, ne saurait fournir beaucoup de variantes intéressantes. La copie elle-méme est loin d’étre correcte. A partir de f. 208, elle devient assez négligente.
= ~
4 aa ay i i a 4
30 Le Codex Q.v.I. 6-10 de Léningrad
Peut-étre aurons nous |’occasion de publier ailleurs les variantes de notre MS. dont nous avons dressé la liste. Mais cette liste, étant trop longue, aurait encombré le présent article. Sans parler de la haute ancienneté de ce MS. qui donne une valeur particuliére 4 son texte, il y a des variantes qui ont une importance absolue. I] est évident que le Leninopolitanus du vr‘ siécle ne serait point 4 dédaigner dans une édition critique du texte.
§3. Les MARGINALIA
Mais ce qui est non moins intéressant dans le MS. (et importe en- core plus dans le plan de cette étude) ce sont ses marginalia, qui jet- tent une lumiére sur ses origines et ses destinées. Nous allons les citer dans leur ordre chronologique, établi d’aprés l’examen de I’écri- ture.
Tout d’abord celles qui, 4 notre avis, sont du méme siécle que le MS., bien que le Nouveau Traité les date du vit siécle (111, 155, note). D’époque trés reculée doit étre la glose du f. 113 au dessous de l’explicit ot l’on ne peut déchiffrer presque rien: ex . . . tudae(?) die uero(?) .. udis tui dicit dns ad populum/ sub sacerdot ..(?) populi... publicum/ .. . illos . . [con] firmans .. . (v. Planche 1 de notre Catalogue). D/’autres se présentent en onciale, en semi-onciale, et en cursive romaine, inscrites en marge des passages (= Pair. Lat., XXI, coll. 1131, 1132, 1144, §§ 17, 19, 38). Ce sont: f. 14v.: de ori/ge/ n/e (onciale); f. 17: de sol et luna et stellis (cursive romaine); f. 39: sollicite lege/uide/ quid lateat (semi-onciale). Cette glose-ci est en marge du texte: Quod liberi pro parentibus non puniantur nec paren- tes pro liberis, ou Vannotateur (cf. ci-dessous, p. 31) a senti, sans doute, le ‘venin pélagien.’ !
Du vir siécle, ou du vir* peut-étre, seraient, en marge du traité de Fulgence, les gloses en cursive qui indiquent les hérésies visées par le traité: f. 64: contra arium; f. 65: contra fotinianos uel mani- chaeos; f. 68: contra apollinaristas; f. 70: contra nestorium; f. 71: contra euticianos; f. 72: contra pelagianos; f. 78: contra marcionitas et alias pestes huiusmodi similes.
1 Cf. Analecta medii aewi, Fasc. mt, ‘Les anciens manuscrits latin de la Bibliothéque Publique de Léningrad,’ Léningrad, 1929.
’ = 4 |
sicem AMdjSuBs TANT NATU ses
eval 1 sps S&C IFATCNTECSUN
TRINTTATISE SS CUIRTUTE RD FURASPATRES me peat ore sone” Re /
web
[NEAT ONT AMER
PuancuE II a
(Toc. MS. Lat. Q. v. 1. 6, fol. 61r
7 lz | | 4 5 4 | - | 1: ue 4 | | | | |
a >
RTASY URN EXSIN UMSICPOs s sin}.
pe ACE RE
P
_ DNede-
PLancue II
LEnrncrad, Pusuique (Toe. Bu6.1-ka), MS,
— veo £8
Lat. Q. v. 1. 8, fol.
PURI
ui | a > ij ~ Midi q >
q
Le Codex Q.v.I. 6-10 de Léningrad 31
Mais, en dépit du Nouveau Traité (cit. supra), nous attribuons au vi° siécle (conclusion que, vu son importance, nous allons examiner i part, p. 34 suiv.) une inscription, qui se trouve en marge supé- rieure du f. 1’. Déchiffrée jadis par les auteurs du Nouveau Traité (11, 155, note), trés effacée depuis, récrite en partie, elle contient des mots qu’on ne lit plus. Il y a d’autres que nous croyons avoir lus d’une fagon plus exacte, comme on pourra vérifier sur notre Planche 1.
hic liber qui attitulatur rufini non te seducat o pie lector/ quia pela- gianus est et blasphemiis pelagianorum plenus/ simulans enim contra arrianos disputationem uenena suae/ haereseos inseruit unde hortor cari- tatem tuam ut hanc/ blasphemiam de uestro codice abscidatis et pro ea librum sci/ augustini de uera religione describite ut quantita/tem codicis reparetis.
Nous croyons devoir attribuer au vir‘ siécle un fragment trés effacé, écrit en cursive ‘barbare’ au verso du f. 220 (qui est le dernier de tout le codex). Le Nouveau Traité (11, 416) donne la lecture de deux lignes du milieu: . . . cane una uinum pfer] totum estu[m] per singulas diae flascune una / alaeo laride pane quantum d{eu)s scit . . .
On peut déchiffrer aisément, ga et 14, encore un nombre de mots isolés. Nous nous abstenons de les citer ici, tant que nos efforts pour identifier le texte intégral (qui est sans doute un fragment ou une paraphrase d’une régle monacale) n’ont pas abouti 4 quelque chose de certain. En attendant, un érudit versé dans la littérature des régles le fera, peut-étre, avant nous, d’aprés le facsimilé que nous donnons ci-dessous sur la Planche v.
Il faut sans doute attribuer 4 l’époque précaroline: vir‘ siécle, a un scriptorium franco-gaulois, deux gloses (Planches 1 et 1m) dont la seconde, celle du f. 61, est trés caractéristique de la main qui a con- servé les habitudes ‘mérovingiennes.’ L’autre qui lui est symmétri- que est trés corrompue, se trouvant récrite sur les traces de l’ancienne écriture. La parenté intime de ces deux gloses nous oblige a les attri- buer a une seule et méme époque et a les considérer comme une appli- cation du conseil donné par la glose du vi‘ siécle: f. 1, cursive: hic liber non est rufini sed / pel{agii] her{etici] in quo contra fidem chato- licam mul/ tae continentur blasphemiae; {. 61, au dessus de l’explicit
| a4 Pit | i i .
32 Le Codex Q.v.I. 6-10 de Léningrad
qui est rayé: hucusque blasphemiae pelagii heretici contra fidem catholicam/ quas sub nomine rufini catolici falso titolo indidit.
C’est a l’époque caroline (vimm*-1x° siécles) qu’il faut attribuer les signes de ponctuation qui modifient le systéme original.
Du 1x/x est Dincipit du f. 115 v.: inctp[it] tractatus primus in libro/ cantict canticor[um] origenis que[m] ieronimus/ de greco in lati- num transtulit.
Du x/xr* siécles est l’incipit, surajouté devant le traité Ad Fabi- olam, f. 114, minuscule: tractatus de quadraginta mansionibus filio- rum israel. _
Du x1/xu® siécle est sans doute le sommaire en minuscule ‘gothique’ inséré vers la fin du MS., f. 220: liber cuiusda{m] heretici pelagu de fide quem sub nomine rufini edidit./ eplistojla fulgentit ep[iscop)t de fide catolica./ origenis libri. 11. in canticum canticorum./ tractatus de .XLII. mansionibus filiorum israel./ epistola s{an)c[t}t hieronimi ep{iscop}i ad demetriadem uirginem.
Au xur° siécle nous attribuons inscription du f. 1: fulgentius de fide catholica.'
Le xvii‘ siécle a inséré plusieurs annotations dont les plus im- portantes au f. 1 sont qui suivent: Pelagii sub Rufini no[m)i[nje errores. En bas: 580 (le numéro du catalogue de St-Germain de 1677).2 En marge: Corbeiensis monasterit et en bas: Editus a Sir- mondo, recusus a Garnerio inter libellos pelagianae confessionis.
Au xvii siécle, on a inscrit: N 840 (du catalogue de 1740). De la fin du méme siécle, aux ff. 1, 61, 62, etc.: Ex musaeo Petri Dubrowsky 1792, et aux folios du papier, les références de Doubrovsky au Nouveau Traité.
Du xix‘ passim: annotations des bibliothécaires de la Lénino-
politana.
1 Se trouvant au premier folio, l’inscription, qui n’a rapport qu’au 2° traité, para‘t étre dé- placée. Elle exprime, pourtant, peut-étre, une idée du contenu du MS. qui s’est formée par suite d’un examen superficiel. Cf. ci-dessous, p. 33.
2 Leopold Delisle, Le Cabinet des Manuscrits de la Bibliothéque Impériale (Paris, 1868), 11, 189; B. de Montfaucon, Bibliotheca Bibliothecarum Manuscriptorum Noua (Paris, 1739), 11, 1133.
| i s n tr ti ci le ve nu tri
Le Codex Q.v.I. 6-10 de Léningrad 33
§ 4. Inpications INTERIEURES suR L’Historre pu MS.
L’inscription du xvii‘ siécle atteste la présence du MS. (voir Planche 1) 4 Corbie. Quand y est-il entré? Les notices des cata- logues des x1* et x1mr° siécles (Delisle, op. cit., 1, 430, No. 140 et 437, No. 173), ott Staerk cherche une mention de notre MS., semblent, au premier abord, n’avoir rien a faire avec lui. Ce8 indications parcimo- nieuses: Fulgentius De fide catholica, et, tout simplement: De fide ca- tholica — saurait-on les identifier avec le riche contenu de notre MS. ou quatre auteurs célébres sont représentés?
Et, tout de méme, l’hypothése n’est pas invraisemblable. Au con- traire. Comme au f. 1", quasi en titre de tout le MS., on trouve l’in- scription: fulgentius de fide catholica (Planche 1; cf. p. 32), il est évident qu’a un certain moment de la vie du MS., la confusion se produit: on a designé ainsi le MS. en simplifiant son contenu. Une erreur analogue pouvait se produire dans le catalogue. La source du malentendu aurait pu étre identique.
Tl est donc probable que, 4 une époaue trés reculée, le MS. se trouva i Corbie. Mais alors une question se pose: n’est-ce pas 14 qu’il fut écrit? Ce qui aurait modifié d’une fagon décisive l’attribution chron- ologique, Corbie n’ayant été fondée que vers le milieu du vit siécle.
E. H. Zimmermann! le prétend. N’ayant — chose incompréhen- sible — examiné le MS. qu’a partir du No. 7: le traité de Fulgence (les nos 6 et 10 ne se trouvent méme pas mentionnés par lui), et s’'appuyant surtout sur l’examen des dessins et des couleurs qu’il caractérise comme corbéiens: sur ceux surtout de l’explicit de Fulgence, il at- tribue le MS. a l’atelier de Corbie, et par suite, au vu‘ siécle. Il se trouve ainsi en contradiction avec le Nouveau Traité (avec Staerk également) et avec l’aspect, trés ancien et trés italien, du MS. (cf. ci-dessus, pp. 23 suiv.).
Pour la plupart des vignettes (dessins: spirale, trés légére; cou- leurs: jaune et rouge, assez délicats) et des initiales méme, avec leur vert qui est bleudtre, on ne saurait dire si elles appartiennent aux nuances spéciales 4 Corbie, que Zimmermann a décrites avec mai- trise. Quant aux dessins, ils s’accordent trés bien avec |’attribution
1 Vorkarolingische Miniaturen, Texte, p. 201, Pl. 113.
i ay | t i i
34 Le Codex Q.v.I. 6-10 de Léningrad
italienne (cf. p. 24, ci-dessus), et méme s’opposent 4 celle de Corbie (cbid.).' Il n’y a que le cadre de l’explicit de Fulgence? (f. 113, Planche Iv) qui a vraiment l’aspect trés corbéien par rapport a la couleur et au dessin, et qui, étant unique, déroge 4 ’harmonie de la décoration. Devant une contradiction sans issue, ne doit-on pas avoir recours 4 un moyen héroique? supposer que ce cadre-ci a été peint dans l’atelier de Corbie, mais postérieurement, le reste de la décora- tion étant ancien? Ce qui est d’autant plus vraisemblable que la glose ‘mérovingienne’ du f. 61 (v. Planche 1) fait penser que le MS. est parvenu au nord de la Gaule vers la fin du vit siécle. Quant a son berceau, nous reviendrons a cette question aprés avoir examiné le probléme de la date de la glose du f. 1’.
En 1638, au nombre d’autres MSS corbéiens, il est entré a la maison de St-Germain et se trouve indiqué par le no. 580 (Planche 1) dans l’ancien catalogue de cette bibliothéque: celui de 1677 (Montfaucon, op. cit., 1, 1133). Il est numéroté ensuite par le no. 840 (inscription au f. 1, Planche 1) qui est celui du catalogue de 1740. Enlevé en 1791, comme Il’atteste la liste de Dom Poirier (Delisle, op. cit., 1, 55), il a été vendu 4 Doubrovsky qui a mis ses notices la- dessus (cf. p. 22). Sur sa division en cing volumes voir ci-dessus page 22.
§ 5. La Date PROBABLE DE LA GLOsE Fotto I‘
I] serait d’une importance particuliére de pouvoir dater cette glose (voir Planche 1) qui, malheureusement, 4 cause des retouches pos- térieures, a, en partie du moins, perdu son caractére authentique. Le fond original, pourtant, lui est resté.
Il faut faire abstraction de toutes les lettres (particuliérement nombreuses dans les deux premiéres lignes; exemple: l’e ‘mérovin- gien’) qui ont été récrites 4 une époque incertaine. La main qui a retouché ce fragment parait étre la méme qui a résumé son contenu en marge droite, 4 cété.* Les formes et l’encre (roussftre) sont ana-
1 Disons entre parenthéses: tout d’abord cette attribution a provoqué des doutes dans la section occidentale du Département des MSS. Exprimés par mon éléve M. Bakhtine, ces doutes me furent une raison de plus pour examiner le MS. plus attentivement.
2 Tl ne donne (Taf. 113) que le facsimilé de cet explicit. De méme ce n’est que celle-ci qu’il mentionne sur cing planches données par Staerk. Cf. Zimmermann, op. cit., p. 201.
3 Cf. ci-dessus, p. 31.
SO
im F
Le Codex Q.v.I. 6-10 de Léningrad 35
logues. Donec, le fragment aurait été retouché 4 la fin du vit siécle. Les lettres plus récentes éliminées, le fond de l’écriture se révéle plus pale.
En examinant ce résidu, nous avons eu l’occasion d’appliquer les observations de W. M. Lindsay (‘The Letters in Early Latin Minu- scule,’ Palaeog. Lat., 1). Cette occasion, pourtant, s’est présentée moins souvent que nous |’avons espéré. C’est que les recherches de M. Lindsay concernent plutét les régions transalpines et |’époque qui, semble-t-il, est postérieure 4 notre glose.
Ce qui saute aux yeux dans notre glose, c’est, qu’en dépit de formes cursives (on n’y trouve ni I’a oncial, ni le N capital, les d et b étant toujours cursifs, excepté un d dans le mot disputationem), les allures générales en sont ‘onciales.’ Les mots ne sont pas séparés: les lettres, au contraire, le sont, sauf les cas de lettres conjointes: les a et e avec leurs voisines (as, er, en, em, a; Planche 1, lignes 1, 3 et 4). L’a est fermé, la haste droite inclinée 4 gauche. L’arrondissement du b al’aspect d’un ceuf élargi en haut, serré en bas. Le c monte au dessus de la ligne, mais il est écrit d’un seul trait (cf. Lindsay, art. cit., §14). L’e, partout ow il n’est pas récrit, a l’aspect trés ancien: e’est un e oncial, incliné et parfois trés allongé (cf. Lindsay, art. cit., §§ 26 et 27). Le g dans les mots pelagius, pelagianus, augustinus, religio a des formes qui varient chaque fois (v. lignes 2 et 6). Leh est arrondi légérement en bas. Le q frappe par son arrondissement, de forme disgracieuse et 4 peine collé 4 la haste. Les r et s sont trés énergiques et le ¢ n’a qu’une seule forme, ‘grecque’ (7).
Ces formes parfois si archaiques, le caractére général des ‘lignes pales’ du fragment feraient attribuer la glose 4 l’époque trés reculée. Une forme, pourtant, inspire des doutes. C’est l’a fermé. On sait qu'il est ouvert dans les cursives des vi—v1i siécles.
En quéte d’analogies, nous avons essayé de confronter les ‘lignes pales’ de notre glose avec les marginalia de saint Augustin du vi‘ siécle 1 et méme avec celles de Térence (le Bembinus).? L’analogie
1 Paris. 12214 + Leninopol. Q.v. I 4. L. Delisle, op. cit., rv, Pl. v1; Emile Chatelain, Un- cialis Scriptura Codicum Latinor. (Paris, 1901), Pl. xxx11 et p. vii; Staerk, op. cit., 11, Pl. xx.
2 Le Vaticanus 3226. Cf. E. Chatelain, Paléographie des Classiques Latins (Paris, 1884- 1900), Pl. v1; Zangemeister-Wattenbach, op. cit., Pl. vi11; Palaeographical Society, Pl. 135; C. Wessely, Schrifttafeln, etc., No. 37; Hermes, 1, 27; M. Prou, Manuel de Paléographie. Recueil de fac-similés décritures du XII* au XVII° siécles (Paris, 1892), p. 49.
| 1 & | le | 4 f t i a il if {
36 Le Codex Q.v.I. 6-10 de Léningrad
avec celles-la est souvent parfaite. Le ductus des scholies en marge du Bembinus est sans doute différent. Mais ce qui est curieux, c’est que les formes isolées sont trés proches 4 celles de notre glose. Dans ces scholies, qui sont généralement attribuées au v—vi siécles, |’a est fermé. L’arrondissement du b est exactement de la méme forme que chez notre anonyme. Le c est trés haut, écrit d’un seul trait. Le d est cursif, le h est arrondi en bas. L’e (de forme onciale) est conjoint & ses voisines. Le q a la forme disgracieuse de notre glose et le ¢ n’a que la forme ‘grecque.’ Les mots ne sont pas séparés et les lettres le sont. Postérieure sans doute a l’époque des scholies, proche a celle des marginalia du Q. v. I 4, notre glose peut étre attribuée a l’époque de l’antiquité en déclin: la ‘Spiitantike.’
Et ceci d’autant plus que d’autres indices soutiennent cette date. Est-ce qu’on pourrait attribuer au vir‘ siécle une glose si correcte? A citer le mot énoncé par L. Traube a un autre propos: ‘propter orthographiam adhuc incorruptam infra saeculum vi extremum de- scendi non potest.’
Que l’on prenne, par exemple, un échantillon en semi-cursive que Steffens date du commencement méme du vir‘ siécle (F. Steffens, Lateinische palaeographie, 2™° éd., Tréves, 1909, Pl. 22). C’est une liste des saints de Rome dont les lampes des reliques ont fourni 4 un certain clerc Jean l’huile pour la reine Théodelinde. La suscription de Jean qui atteste avoir recueilli ces huiles temporibus domni Gregorii papae (voir la planche citée de Steffens), fait attribuer le document a l’époque aprés, mais peu de temps aprés, la mort du Pape. Dans cette liste avec ses cum tres filias suas (col. 1, |. 5), Pauli apostholi (ll. 1 et 2), Cornilit (1. 8), ete., la confusion, la bar- barie fait déja son avénement. Tout est, au contraire, irréprochable dans notre glose. Non seulement un bon latiniste se révéle dans ce texte, mais un grécisant, dans la forme du mot haereseos. En outre, ce doit étre un écrivain de l’époque reculée, et en tout cas antérieure au vir* siécle qui ne connait point d’autres abréviations que sc?.
Enfin: le contenu de la glose fait penser qu’elle ne saurait étre de beaucoup postérieure au texte foncier. L’angoisse d’une Ame ortho-
1 L. Traube, Mon. Germ. Hist., Auct. Antig., x11, 462.
] I b P
Le Codex Q.v.I. 6-10 de Léningrad 37
doxe en face du texte soupcgonné de l’hérésie pélagienne, est-elle du vir‘ siécle?
L’attribution de la glose au vr‘ siécle s’impose avec une grande probabilité.
§ 6. Les OricINEs pu MANUuScRIT
Qu’il nous soit permis maintenant d’ajouter quelques considéra- tions plus hypothétiques concernant les origines du MS. précieux.
Quatre annotateurs du MS., ceux des vi‘, et siécles (cf. les inscriptions), reviennent au nom de Pélage, inspirés sans doute par l’autorité du premier qui a suggéré ce nom. Les érudits des trois derniers siécles ont rejeté cette hypothése.! I] n’est pas de notre intention d’y revenir. Mais une autre question pourrait étre examinée: le premier annotateur, oi a-t-il puisé cette idée? Ce nom, comment lui est-il venu? Enfin, qui était-il lui-méme? Une ré- ponse plus ou moins certaine 4 ces questions ne peut étre déduite des données dont nous disposons. On peut pourtant essayer de re- stituer le milieu dans lequel le MS. a surgi, tracer un cercle plus étroit autour de la personnalité de l’annotateur anonyme.
On reléve dans notre codex nombre de traits qui raménent la pensée vers les instructions données jadis aux bibliothécaires et copistes de l’Occident par ‘le dernier des Romains et le premier des hommes du Moyen Age’:? le fondateur du Vivarium, Cassiodore. Notre recueil contient deux traités contre l’hérésie, deux homélies, deux épitres. Ne dirait-on pas que c’est 1a l’exécution du programme des lectures pieuses, proposé par Cassiodore, programme qui, en méme temps, peut étre considéré comme un renseignement sur le contenu de la bibliothéque rassemblée par lui pour l’instruction de moines du Vivarium? * 1 Cf. J. Sirmond, op. cit. supra, Préface (Lectori); D. Vallarsi, dans Patr. Lat., xx1, 16 et 294; Herzog-Hauck, Realencyclopédie, article ‘Pelagius’ par Fr. Loofs: ‘Ersteres Werk (Pelagii
de fide trinitatis) bis jetzt spurlos verloren’; H. Zimmer, Pelagius in Irland (Berlin, 1901), pp. 13 et 15; Otto Bardenhewer, Geschichte der altkirchlichen Litteratur (2° éd., Freiburg im Br.: Herder, 1912-1924), 111, 558.
2 Feodor Schneider, Rom und Romgedanke im Mittelalter (Munich: Beck, 1926), p. 90.
3 Le fait est attesté par le De institutione avec ses: reliqui uobis, uobis dereliqui, habeat bibliotheca uestra . . . (coll. 1112, 1114, 1115, 1126, 1130, 1134, et passim); cf. surtout le premier chapitre. Les recherches des prédécesseurs de Rud. Beer (A. Olleris, Cassiodore, Con-
-B | ; a
38 Le Codex Q.v.I. 6-10 de Léningrad
Epistolae quoque patrum, sermones et libri diuersorum nec non et homiliae uel cum haereticis altercationes fidelium magno studio relegantur (De instit. diuin., Patr. Lat., uxx, 1130).
Pour choisir sa matiére, le rédacteur de notre recueil s’adresse a celui qu’il croit étre Rufin, au traité de Fulgence contre l’hérésie, au commentaire d’Origéne sur le Cantique des Cantiques, aux épitres de saint Jéréme. Du reste, Origéne est offert dans Ja version de saint Jéréme, avec l’épitre introductoire de celui-ci, adressée au Pape Damase.
On dirait que c’est encore le génie de l’auteur de De institutione qui inspire ce choix. I] reeommande beaucoup la lecture de Rufin et la copie de ses ceuvres, versions et commentaires (ibid., coll. 1117, 1121, 1125, 1133).1. Tout méfiant qu’il est vis-a-vis d’Origéne, quem multorum ... sententia designat haereticum (ibid., col. 1111), tout de méme, plein d’un intérét passionné pour lui, il se sent rassuré par le fait que sanctus Hieronymus etus aliqua opuscula sermone disertissimo transtulit in latinum (ibid.). Tl revient constamment a son ceuvre au cours du traité De institutione (coll. 1111, 1112, 1113, 1114, 1117, 1120). Ses sympathies pour les homélies d’Origéne In Canticum Canticorum sont d’autant plus exemptes des inquiétudes d’une conscience catholique que, traduites et reeommandeées par saint Jéréme, expliquées par Rufin (cbid., 1117), l'un et l’autre sont garants de l’orthodoxie du poéme extatique. Celui-la plus particu- liérement. L’admiration de Cassiodore est sans réserve pour saint Jéréme: Planus, doctus, dulcis, parata copia sermonum ad quam- cunque partem conuertit ingenium (ibid., col. 1135), ce sont ses épitres surtout qu'il apprécie. Copiosas epistolas fecit: beati quibus scribere ... dignatus est (ibid.). Il y goate des nuances de cceur et de style, en insistant sur le culte que saint Jéréme a eu pour la virginité: servateur des livres de lVantiquité latine, Paris, 1841; A. Frantz, M. Aurelius Cassiodorus Senator, Breslau, 1872; cf. L. Traube, Vorlesungen und Abhandlungen, 1, 107) sur cette biblio- théque, ainsi que ses recherches a lui, sont résumées par R. Beer dans les ‘ Bemerkungen iiber den iltesten Hss. bestand des Klosters Bobbio’ (Acad. de Vienne, Anzeiger, hist.-phil. Klasse, Mai, 1911, pp. 78-104): ‘Cassiodorus die in seinen Werken herangezogenen Schriften nicht bloss nennt, sondern in den meisten Fiillen vor sich hatte, ja zumeist ausdriicklich versichert, dass er sie der Studienbibliothek der Ménche in Vivarium einverleibt hatte’ (op. cit., p. 83).
1 ‘Dass Rufin einer der von Cassiodorus am meisten geschiitzten Autoren war, ist bekannt’ (ibid., p. 98).
4 } ek 4
_ TUR Gu amMopor LOAd SeCTRADET. UTdESCCNDE REOPACTAT AD MNEDECORANDAC Ret TE alo emases Ssponsam: QMAceNAetuac GMaManas quiap edestur SSICONC ESSE RIT SpPrrsr ayo OOSpons ACNE AUA RIE STN: cularu Mparnan yLaus wNemen AMAONO STRAAM VOC siomlirie RLABROREMm Us,
PLancueE ITI
Pusiique (Moc. MS. Lat. Q. v. I. 8, fol. 128r
| | | | . | | | q i ug 4 | | | Ny | q | K e
c r d t d a Fe ci
Le Codex Q.v.I. 6-10 de Léningrad 39
modo uirginitatem praedicans ... (tbid.). Qu’on se souvienne que la belle Epistola ad Demetriadem Virginem, apologie ardente aeternae pudicitiae (notre codex f. 193 = CSEL, tv1, 182), achéve et couronne notre recueil.
Ce n’est que Fulgence seul qui n’est pas nommé expressis uerbis dans le traité de Cassiodore. N’y est-il pas, pourtant, visé, lui qui a été l’Ame altercationum fidelium cum haereticis? (ibid., col. 1130). En somme, plus on examine le De institutione en le confrontant avec notre codex, plus il devient difficile d’échapper a l’idée que ce sont les prédilections du Maitre qui ont inspiré le choix du recueil, jusqu’a ce que Cassiodore nous apparait presque comme un génie invisible de la rédaction.
N’est-il pas également celui de l’exécution?
On sait quels soins minutieux le fondateur du Vivarium donnait aux moindres détails de la confection des livres de son atelier.!. La fidélité 4 ses prescriptions n’est pas caractéristique sans doute que pour le Vivarium, devenue la loi de tous les scriptoria qui l’imitérent. Il faut pourtant concéder que c’est au Vivarium qu’il faut penser, plus particuliérement du vivant du maitre, avant que l’orage longo- bard ait éclaté. Quelles sont les observations suggérées 4 cet égard par notre codex?
Cassiodore appréciait beaucoup et recommandait a ses disciples la netteté, le dessin énergique comme loi de la disposition extérieure: ce qu'il appellait le “démembrement’ du texte (col. 1129). Il admirait saint Jéréme qui disposait son texte colis et commatibus, quoniam illustrem et planissimam faciunt orationem, quando suis locis aptata resplendent (ibid., col. 1129).2 Or nous avons vu ci-dessus (p. 25 suiv.) que les tendances 4 ‘démembrer’ le texte, en l’entourant d’espaces libres ou les locutions puissent ‘respirer’ librement, le sys- téme des alinéas fréquents, des grandes marges auprés des citations, des lignes bréves, formant des colonnes — que toutes ces tendances animent les copistes de notre MS. (cf. Planches u et m1).
1 De institutione, passim; A. Olleris, op. cit.; F. Schneider, op. cit., v1° chap.: ‘Cassiodore’; Fr. Milkau, Zu Cassiodor, Berlin, 1928 (Sonderabdruck aus Von Biichern und Bibliotheken. Festschrift fiir Ernst Kuhnert).
2 “Meminisse . . . debemus . .. Hieronymum omnem translationem ... propter simpli- citatem fratrum colis et commatibus ordinasse’ (col. 1124; cf. col. 1109).
i ‘ | ty t
40 Le Codex Q.v.I. 6-10 de Léningrad
Mais, en dehors de ce procédé, Cassiodore croyait utile un autre (praeter translationem S. Hieronymi quae colis et commatibus ordinata consistitit): le systéme des distinctiones: points qu’il voulait petits et élégants: puncta breuissima pariter et rotunda et planissima (ibid., col. 1129), etc., nam si corpus noster indiget per membra cognosci, cur lectio, cum suis partibus uideatur esse distincta, confusa relinquitur? (col. 1129). Le systéme de nos copistes, nous l’avons vu (p. 25, ci- dessus), se trouve en harmonie avec ces prescriptions.
Les problémes de la grammaire préoccupent Cassiodore au cours de De institutione, surtout par rapport aux noms hébraiques: géo- graphiques et personnels. Et s’il est si éloquent 4 déconseiller 4 ses disciples la méthode erronée de les décliner: hebraea uero nomina hominum uel locorum nulla declinatione frangatis: seruetur in eis lin- guae suae decora sinceritas (col. 1127), il faut constater que les copistes de notre MS. ne semblent avoir péché contre cette prescrip- tion que trois fois: ramassem, maram et manassem, ff. 158, 161, et 190). En examinant le traité De mansionibus filiorum Israel (ff. 153-192), ou: l'occasion d’appliquer cette régle ou de l’enfreindre se présente 4 chaque pas, on a presque envie de se demander si ce n’est pas en vue de ce traité que la régle a été énoncée? En tout cas les sectateurs de Cassiodore s’y sont montrés pleins de zéle, de respect pour la decora sinceritas linguae.
D’autre part, que l’on étudie notre MS. au point de vue de I’or- thographe. On y saura relever certainement (chez le dernier copiste plus particuliérement) des infractions aux régles que préconisait le maitre. Mais on ne peut s’empécher de penser que ce sont justement les peccadilles, propres aux copistes de notre MS., qui ont servi de matiére 4 ses observations. Le MS. ne semble présenter que celles que le chef du Vivarium connaissait si bien.1 En méme temps il faut constater que les erreurs sont relativement rares, et que la copie est plutét correcte.
Ae in fine aduerbii non relinquas, ae iterum casui genitiuo non subtrahas ... (col. 1129). Cette prescription-ci est observée sans faute 4 travers tout notre MS. sauf un seul ‘male’ au lieu de ‘malae’ (f. 163).
1 *. . . nunc dicimus in quibus litteris sunt librariorum uitia corrigenda’ (ibid., col. 1123).
|
|
]
t
q
P
(
L
ff
D
tr
ed
fa
co
per
191
regu
Le Codex Q.v.I. 6-10 de Léningrad 41
Quod, cum pronomen est, per d, non per t litteram, cum uero ad- uerbium numeri est, per t litteram, non per d scribendum est (ibid.). Nous avons relevé quelques cas ow |’on péche contre cette régle. Au f. 212 le copiste écrit guod ( = quot) oris et au f. 113, si quit (= quid) aliter sapit. Les erreurs de cette espéce sont pourtant peu nom- breuses. Et d’une facon générale tout se présente bien avec les d et ¢ dans notre MS., si l’on fait abstraction de plusieurs capud et re- liquid chez le dernier scribe, et aliut et istut chez l’avant-dernier. C’est aux spécialistes de discerner si la source de la confusion est la méme pour le torridam (f. 160 = turritam, CSEL, wv, 56, 1. 5).
B pro v, v pro b, o pro u uiciose positas non relinquas (ibid.). Ces lettres préoccupaient Cassiodore d’une fagon tout a fait particuliére. La place qu’il leur donne dans son recueil de textes des orthographes est considérable, et refléte l’intérét qu’il leur témoignait.!’ Pour en expliquer l’importance, il importune beaucoup les Manes d’Adaman- tius Martyr.’ Si l’on examine notre MS. a ce point de vue, on verra que la confusion augmente plus on avance: parbuli, renobantur, pugnauit (futurum), temptabit (perfectum), uinas (= binas), nobi (= nout), uaculus (= baculus), uellum (= bellum), pleue (= plebe), etc. (resp. ff. 155, 157v, 160v, 161, 162, 172, 174, 175, ete. = CSEL, Lv, 50, 1. 4; 53, 18; 56, 18; 57, 11; 58, 12; 68, 16; 70, 17; 75, 2; 76, 16). La confusion des u et o est un phénoméne beaucoup plus rare dans notre MS., et les formes comme epistula et consolatum (consulatum, ff. 62 et 194) sont exceptionnelles.
Aspirationem uero superfluam deme aut adice competenter (ibid.). De nos copistes les deux derniers méritent quelquefois le reproche de ‘Yincompétence.’ Des halii ( = alii), hora ( = ora), hisdem ( = tisdem), trouvent leur pendant dans des ordaceis (= hordaceis), aut (= haud), edus (= hedus), etc. (resp. ff. 184, 167v, 153, 178, 204, etc.). Ces fautes sont pourtant assez rares, et l’ouvrage des trois premiers copistes en est exempt.
1 “Ad amantissimos orthographos discutiendos anno aetatis meae nonagesimo tertio . . . perueni’: W. S. Teuffel, Gesch. d. rimischen Litteratur (6™° éd., Leipzig: Teubner, 1910- 1916), 111, 496; ‘Ex xu orthographorum libris rectitudinem scripturae docentium defloratas regulas posteris legendas exhibui’ (Gram. lat., v1, 209).
2 Gram. lat., vir, 165-199.
| . q | i" | ij
42 Le Codex Q.v.I. 6-10 de Léningrad
Multa etiam respectu euphoniae propter subsequentes litteras pro- babiliter immutamus, ut illuminatio, irrisio, immutabilis, impius, im- probus (ibid., col. 1129). Les scribes de notre MS. se sont montrés parfaits en ce qui concerne |’observation de cette régle. Contraire- ment a l’usage qui a prévalu depuis et qui s’est reflété: (1) dans les corrections carolines du MS. méme; (2) dans les MSS du rx* siécle qui se trouvent a la base de l’édition de CSEL; donc (3) dans cette édition, — nos copistes écrivent presque sans exception: ammirantur, imminutio, commiscuit, assumans (ff. 158, 166v, 165, 159v, etc.). La méme loi sans doute les incite 4 écrire: impleat, compositum, compen- sabit (ff. 166, 169, 175, etc.). Et bien que tous écrivent d’une facon systématique: scribtum, scribtura, etc., et que ceci ne semble pas étre prévu par Cassiodore, nous demandons aux spécialistes: cette orthographe-ci serait-elle condamnée par lui?
Résumons: sauf le dernier copiste, notre MS. est assez correct. Si Yobservation de Traube: ‘propter orthographiam adhuc incorrup- tam infra saec. vI extremum descendi non potest’ ne saurait étre appliquée a notre MS. qu’avec certaines réserves, s’il n’est pas libre d’erreurs, ce sont, en tout cas, celles au milieu desquelles Cassiodore a vécu, qui lui étaient familiéres. De cette dégénérescence radicale ou, si l’on veut, de cette vie nouvelle de sens et de formes qui se fait jour dans les manuscrits du vit siécle, il n’y a pas de traces dans le nétre.
Il y a dans les souvenirs littéraires de Cassiodore un point un peu obscur. En tout cas, au chapitre xxvi (col. 1140), De notis affigendis, nous n’avons compris que ceci: en relisant au déclin de sa vie (senex aut longa peregrinatione fatigatus) les volumes de sa bibliothéque, Cassiodore, pour en faciliter la revue & ses éléves et pour en faire la lecture plus agréable (ut wos labor noster instrueret et qualicunque munusculo studium uestrae sanctitatis ornaret), annotait, a l’aide de minium (presentes notas minio designatas, quae sunt indices codicum, quibusque locis competenter impressi), les loci qu il croyait dignes d’étre relevés.!
1 W. Weinberger, Die Handschriften von Vivarium, 1925 (Miscellanea Francesco Ehrle, tv, 1924), prétend que les lignes initiales de chaque traité faites en rouge sont |’application de ce procédé de Cassiodore et, en méme temps, |’indice de la bibliothéque du Vivarium. E. A. Lowe
a d d ve I’) qr co ge ge de ib Say re] reje Ive-
Le Codex Q.v.I. 6-10 de Léningrad 43
Ces loci, étaient-ce les initia des paragraphes, et le minium servirait-il 4 faire plus nette la division du texte? Etaient-ce les pas- sages ou les mots les plus importants qu’il soulignait 4 l’aide de rouge? Mettait-il ces signes (notae) en marge (les chiffres romains?), en soulignait-il le texte méme? Tous ces procédés, peut-étre, sont sous-entendus par lui? Ensuite, aprés avoir parcouru le volume, quos tamen sub ipso studio transire praeualut, il dressait 4 son début un sommaire: guas in primordits codicum semper ascripst.
D’un pareil sommaire (ce souvenir devait sans doute servir de modéle aux travailleurs de son scriptortum), pas de traces au début de notre volume. Aurait-il disparu avec les folios coupés? La table des matiéres du x1* siécle, au f. 220, n’en est-elle pas une copie? Peut-étre. Quant 4 l’emploi du minium, il est abondant et trés varié dans notre MS., on I’a vu (pp. 23, 24, ci-dessus). Si les chiffres ro- mains en marge de Fulgence et de l’épitre Ad Fabiolam, si le véte- ment rouge des exhortations Firmissime hoc tene, si les mots, faits au minium: HELIA, MOYSES, etc. aux ff. 101, 206, et 207, si le rouge des lignes initiales et des explicits n’a rien a faire avec ]’expédient de Cassiodore, nous ne savons ce que ces procédés peuvent bien vouloir dire. Encore une observation: toujours sur ses gardes contre lhérésie, Cassiodore recommande une lecture trés attentive, comme qui dirait le crayon 4 la main (voir ci-dessous), des traités qui la combattent. Cum haereticis altercationes fidelium magno studio rele- gantur (col. 1130). Ne dirait-on pas que ces gloses que nous avons citées (p. 30, ci-dessus) qui indiquent, un 4 un en marge de Ful- gence, les ennemis contre lesquels le texte est dirigé, sont conformes i esprit de Cassiodore? On reléve dans ses mémoires une allusion & des procédés analogues. Se méfiant d’Origéne, bien qu’il s’efforgat d’y relever des choses vraies et utiles (awrum in stercore quaero, tbid., col. 1112), il n’omettait pas d’autre part d’y marquer les pas- sages suspects: loca quae contra regulas patrum dicta sunt axpnotwv repudiatione signaui (col. 1112; ef. col. 1122). Quels avaient été les
(‘Some Facts about our Oldest Latin MSS,’ The Classical Quarterly, x1x (1925), 197 suiv.) rejette ce criterium, en tant qu’un indice de Vivarium; ainsi que certains autres (par exemple: les majuscules au début de chaque colonne) qui sont l’indice de la haute ancienneté du MS.:
Iv*-ve siécles. Donc ils ne peuvent étre un indice du scriptorium Vivariense fondé vers le milieu du vi? siécle!
a | | | a
+t Le Codex Q.v.1. 6-10 de Léningrad
signes, dont il marquait ces endroits ‘inutiles’ (a4xpjorwy)? II se vante, en tout cas, de le faire d’une facon intelligible, lui qui détestait les énigmes des notes tironiennes: quibus libris, twuante Domino, capitula insignire curauimus, ne in tam necessaria lectione . . . confusa tironis nouitas linqueretur (col. 1117). Notons ceci: sauf les deux gloses, relativement postérieures (cf. p. 31), en notes tironiennes, notre MS. est remarquable par l’absence absolue de ces symboles dans les marginalia.
Le lecteur qui aura confronté le facsimilé de notre MS. avec la Pl. xvu de Chatelain, Uncialis (cit. supra), qui est celle du MS. de Bobbio, devra s’aviser d’une analogie frappante des deux MSS. Mais évidemment, ce rapprochement n’améne point a l’attribution de notre MS. a l’atelier de Bobbio. Le MS. de Prudence, lui-méme ne saurait lui étre attribué. Les deux MSS étant du vit siécle, et la maison de Bobbio ayant été fondée en 614, tous les deux proviennent d’un berceau plus ancien.
L’hypothése, bien sir, n’est pas exclue que, frére jumeau du MS. de Prudence de Bobbio, notre codex aurait, comme lui, séjourné dans cette maison ou, avec le nombre d’autres MSS, il n’a fait que passer d’un fonds plus ancien. On sait que R. Beer (op. cit. supra) en quéte d’une ancienne bibliothéque qui aurait formé le gros de la Bobbiensis, rejette successivement celles du Pape et d’Eugippe, pour conclure a celle, ‘die Cassiodorus Senator in Vivarium gesammelt hatte’ (pp. 82 suiv.). Que l’on refasse la série de raisonnements qui, en 1911, l’ont amené a chercher, dans les collections ‘antecolumbanes’ de la Bibliothéque de Bobbio, Vhéritage du Scriptorium Vivariense: ‘Keine friihmittelalterliche Bibliothek’ — telle est sa conclusion — ‘die Biicherversammlung Cassiodors so deutlich wiederspiegelt, wie der iilteste Handschriftenvorrat des Klosters Bobbio’ (op. cit., p- 104). Ilajoute une observation qui devient singuliérement vivante dés qu’on l’applique 4 notre MS.: ‘Jene ehrwiirdigen Bobbienses selbst, zu Cassiodors Wirksamkeit in Beziehung gesetzt, eine beach- tenswerthe persénliche Note erhalten’ (ibid.).
Dans ce qui précéde il y a eu des rapprochements un peu larges. Certains d’entre eux, tout de méme, ont contribué — nous I’es-
i 0 li at di ve) dic Ab anc (p. Zeit
Le Codex Q.v.I. 6-10 de Léningrad 45
pérons — a resserrer le cercle autour de notre MS. et 4 indiquer le milieu d’ou il a pu surgir.
Ce milieu, n’est-il pas concevable de le chercher dans le groupe des disciples de Cassiodore, adeptes des méthodes indiquées par l’auteur de De institutione?
Avec le méme degré de probabilité on peut tenter la reconstruc- tion des destinées ultérieures du MS. Les recherches de L. Traube ont mis en lumiére les circonstances dans lesquelles les MSS italiens du vi siécle ont passé en Gaule 4 l’époque de Charlemagne.' Notre codex dont la présence 4 Corbie est attestée par le catalogue du x1* siécle s’est trouvé, certainement, beaucoup plus t6t au nord de la Gaule. C’est la sans doute que le cadre en couleurs avec dessins ‘cor- béiens’ a été peint autour de l’explicit du deuxiéme traité. C’est 1a aussi qu’a la méme époque (vuI* siécle), en écriture ‘mérovingienne,’ on a renouvelé aux premier et dernier folios du premier traité l’aver- tissement de l’annotateur anonyme. Lors de la confection du cata- logue au xI* siécle, on a dressé au f. 220 la table des matiéres qui revient toujours 4 cet avertissement.
Celui-ci regoit une valeur tout 4 fait particuliére sur le fond des observations recueillies ci-dessus. On peut réduire au nombre de cing les théses sur lesquelles insiste l’annotateur anonyme, tout en discutant l’attribution du premier traité 4 Rufin.
1. Ce traité est celui de Pélage ou d’un auteur pélagien: Hic liber pelagianus est et blasphemiis pelagianorum plenus.
2. Pour mieux séduire le lecteur, tout en simulant la polémique, Phérésie se cache sous l’autorité d’un nom vénéré: Hic liber qui attitulatur Rufint non te seducat... simulans... contra arrianos disputationem.
3. C’est au moyen de cette ruse que l’auteur caché insére les venins de Vhérésie: wenena suae haereseos inseruit.
1 Les recherches de L. Traube sur ce probléme dans la Teztgeschichte der Regula S. Bene- dicti (2™° éd., revue par H. Plenkers, Munich, 1910) sont résumées dans les Vorlesungen und Abhandlungen du méme auteur: ‘Karl der Grosse hat zielbewusst Codices der Klassiker und andere Texte . . . im Original oder in Abschrift aus Italien in sein Reich schaffen lassen . . .’ (p. 182). ‘Im allgemeinen ist der Weg der, dass alte Hss. aus dem v1. Jht. nun in karolingischer Zeit fortgeplanzt werden, ohne dass irgendwelche Zwischenstufen da sind’ (p. 183). Cf. F. Schneider, op. cit., p. 95.
| |
|
|
L
46 Le Codex Q.v.I. 6-10 de Léningrad
4. Le ‘lecteur pieux’ (ce doit étre un du groupe qui a créé le re- cueil, plutét le ‘groupe’ lui-méme: le singulier tuam passe en pluriel uestro) est supplié d’expurger le recueil de ce traité: hortor caritatem tuam ut hanc blasphemiam de uestro codice abscidatis.
5. Asa place on est invité de mettre le traité de saint Augustin, De uera religione, pour ne pas déroger 4 l’intégrité du volume: pro ea librum s{an]c[t}i augustini de uera religione describite ut quantitatem codicis reparetis.
Encore une fois, la glose fait penser aux aveux de Cassiodore qui, toujours hostile 4 tout ce qui sentait lhérésie de Pélage, prétendant avoir bon ceil pour la discerner,! s’adresse 4 ses disciples avec une exhortation que nous allons citer, tout en mettant en italiques tous les passages dont la coincidence avec la glose anonyme nous a
frappée.
In epistolis tredecim sancti Pauli adnotationes conscriptas in ipso initio meae lectionis inueni quae in cunctorum manibus ita celebres habeantur ut eas a sancto Gelasio papa urbis Romae, doctissimi uiri, studio dicerent fuisse conscriptas. Quod solent facere qui res uitiosas cupiunt gloriosi nominis auc- toritate defendere. Sed nobis ex praecedentibus lectionibus diligenti retrac- tatione patuerunt subtilissimas quidem esse et breuissimas dictiones sed pelagiani erroris uenena illic esse seminata.
Et ut procul a uobis fieret error haereticus, Epistolam ad Romanos qua potui curiositate purgaui, reliquas in chartaceo codice conscriptas uobis emendandas reliqui. Quod facile subiacebit, quando praecedenti exemplo audacior redditur sequentis imitatio (De instit., Patr. Lat. uxx, 1119).
Les théses de l’anonyme se retrouvent, parfois littéralement, dans les aveux de Cassiodore (ou inversement, si l’on veut).
1. Dans les commentaires sur saint Paul, l’arriére-fonds pélagien se révéle: patuerunt.
2. Il se cache sous l’autorité d’un nom vénéré: le Pape Gelase.
3. Ainsi on cherche 4 insérer le venin pélagien: pelagiani erroris uenena illic esse seminata.’
4. Cassiodore s’efforce d’en expurger le volume — purgaui — pour en confier |’exécution définitive 4 ses disciples: wobis emendandas
reliqui.
1 Cf. H. Zimmer, Pelagius (cit. supra), pp. 15 et 200 suiv,
‘
-
ig
is
AI SHONVIG
ANS
Nid SANLVOL AD
AVON INGXIAL
be
A SHONVIg
~
| lel tior De san dor sior dan Au stiq par | not | lau | poll tiqn du les; aal sior pél: met par
Le Codex Q.v.I. 6-10 de Léningrad 47
Il n’y a que le conseil de l’auteur anonyme de la glose concernant le De uera religione de saint Augustin qui manque dans la déclara- tion de Cassiodore. Mais, en effet, on le retrouve encore dans le De institutione:
Utiles etiam sunt ad instructionem ecclesiasticae disciplinae memorati
sancti Ambrosii de Officiis melliflui libri tres, nec non et beati Augustini De uera religione liber unus (ibid., col. 1132).
La liste des ceuvres de saint Augustin recommandées par Cassio- dore n’est certainement point réduite a ce traité qui, par ses dimen- sions, aurait suppléé l’ceuvre suspecte. Cassiodore, ayant indiqué, dans le paragraphe qui précéde, toute une série d’ceuvres de saint Augustin — dans ce paragraphe-ci, qui porte sur la discipline ecclésia- stique, recommande, en premier lieu, le traité mis en avant aussi par notre anonyme.
Résumons.
(1) L’hypothése qui présume que le Vivarium fut le berceau de notre MS. au vi* siécle est trés vraisemblable.
(2) Un groupe de disciples et de sectateurs de Cassiodore en a été lauteur, resp. le rédacteur, du vivant méme du Maitre.
(3) Ils Pont composé d’aprés son programme et l’ont exécuté, au point de vue de la disposition du texte, de l’orthographe, de la phoné- tique, des formes des noms hébraiques, de la méthode de l'emploi du minium (peut-étre), et des gloses indicatrices, conformément a lesprit et 4 la lettre des indications de Cassiodore.
(4) Les erreurs qu’ils y ont commises sont toutes propres a Pépoque de Cassiodore et trés bien connues de lui, étudiées par lui d’une fagon systématique.
(5) Un de ces sectateurs, qui n’a pas pris part 4 l’exécution du recueil, mais qui était imprégné des idées antipélagiennes du Maitre, a annoté le MS. a son début et, en se servant de certaines des expres- sions propres 4 Cassiodore, a condamné le premier traité, comme pélagien.
(6) Il conclut a la proposition de le rejeter tout a fait, pour mettre a sa place le traité de saint Augustin, reeommandé également par le Maitre, au début du paragraphe de la ‘discipline.’
i } | if id
48 . Le Codex Q.v.I. 6-10 de Léningrad
Quelqu’un qui aura examiné et accepté ces raisonnements va, peut-étre, poser une question que (inutile de le cacher) nous nous sommes posée nous-mémes.
Un sectateur, un disciple. ... Et pourquoi pas le Maitre lui- méme?... Senex aut longa peregrinatione fatigatus, ne pouvait-ce étre lui-méme qui, aprés une intervalle, relegere praeualuit ce qu’on a exécuté dans son atelier, pour in primordiis codicis en donner son avis? II le faisait parfois: loca quae contra regulas patrum dicta sunt axpnotwy repudiatione signaut.
Mais alors serions-nous donc en présence de l’autographe de Cas- siodore?
L’hypothése serait beaucoup trop séduisante. L’attribution de la glose du premier folio au vr‘ siécle n’est pas exempte de doutes. La main courante de Cassiodore nous est inconnue. Nous n’avons donc rien qui aurait donné une certitude parfaite. Il vaudrait mieux de nous arréter la. Puissions-nous ne pas étre atteinte des foudres du Maitre, au nombre de ceux qui res uitiosas cupiunt gloriosi nominis auctoritate defendere.
Léninarap, U. R. S. S.
BEATIS | F. 191
PuLancue VI
LéninGrAD, (Toc. MS. Lat. Q. v. I, 6-10, ff. 62, 114, 132, 154, 191
F. 114 id g j aw 4 |
exé (Is in| ext us que | his Ag boc tru me lar! are ver qui line 1 fort, in hi Juss Dev prin’ inde 1ff., is al: LUM, (192
THE BELLUM TROIANUM OF JOSEPH OF EXETER‘
By WALTER BRADBURY SEDGWICK
T WILL be convenient to begin with a short account of the Metre
and Grammar before proceeding to the main business, a detailed examination of the Style of the Bellum Troianum of Joseph of Exeter (Iscanus) : we conclude with remarks on the Text and Vocabulary.
I. Metre
Joseph’s hexameter is carefully modelled on that of the Silver Age, in particular of Lucan (and Claudian)— though he manages to some extent to avoid their monotony: in,his bold use of elision he reminds us of Virgil, though he sometimes — not often — goes too far (e.g. qua acies, vi, 274; boum hasta, vi, 501; cum ad, iv, 506). On the whole his versification is little, if any, inferior to the average of the Silver Age. When we consider the futility of the usual mediaeval hand- books of metre, we are lost in admiration of a poet whose ear was so true that it enabled him to reproduce exactly niceties of the hexa- meter only revealed by nineteenth-century scholarship (particu- larly in respect of elision, e.g. of cretic words *). His chief weaknesses are: (L) The Golden line (two nouns with their adjectives and a verb); in the examples below I have included a few which are not quite strictly such, but have virtually the effect of the true Golden line; (2) verses of the type aequoreum | sulcabat | iter || — which are
' T use the Delphin reprint (London: Valpy, 1825) of the edition of S. Dresemius (Frank- fort, 1620), who claimed to be the first to restore the poem to its true author, as did J. Morus in his edition (London, 1675). But the authorship was known to Leland and Camden (see Jusserand, op. cit. infra, pp. 82 ff.); cf. also Samuel Daniel’s Defence of Rhyme: ‘Josephus Devonius who wrote de Bello Troiano in so excellent manner, and so resembling antiquity, as printing his work beyond the seas they have ascribed it to Cornelius Nepos.’ I am also indebted to Mr R. K. Root’s article on ‘Chaucer’s Dares’ in Modern Philology, xv (1917), 1 ff., and to J. J. Jusserand’s early thesis, De Iosepho Iscano (Paris: Hachette, 1877). Reference is also made to my article, ‘Style and Vocabulary of the Latin Arts of Poetry ...,’ Specu- LUM, 111 (1928), 349 ff. For key to abbreviations employed, see this article and SpecuLum, 11
(1927), 331 ff. ? These are not elided by Joseph and the best classical poets.
49
50 The Bellum Troianum of Joseph of Exeter
characteristic of Lucan; (3) lines with noun and adjective rhyming at the middle and end of a line, an Alexandrian trick, common in Latin from Ovid onwards — carried to excess in some Christian poets, e.g. Paulinus’ Vita Martini (see statistics for all in Max Manitius’ Geschichte der Christlich-Lateinischen Poesie, Stuttgart, 1891) — and leading to the Leonine: it is not very common in Joseph. These three types occur in succession in vi, 8-10:
At (read ac) sonipes animi | non argumenta superbi || (3)
Iratas | iterat uoces ||, non puluere capto (2)
Multiplicat suspensa leui uestigia gyro. (1) There is a certain amount of monotony, especially in the almost inevitable pairing off of noun and adjective, but not more than in say, Claudian, who, for instance, in De Raptu Pros., Book ii, has 19 Golden lines in the last 60 verses.1 But on the whole Joseph’s verse is very fine indeed, especially when compared with that of any other mediaeval poet. Bad verses are very few — e.g. vi, 302:
Argolica: illic ante ipsum mactatus Achillem.
The endings erit una, ii, 285, minus uno, ii, 161, can be paralleled from Virgil. I have noted no other verses as defective (foeda ubi uictus, ii, 327, as an ending will, I think, pass muster).
Like other mediaeval writers he retains a short final vowel before sp, sc, and so forth, and lengthens a short syllable at the strong caesura — but mostly only with Virgilian precedents. (Unlike his contemporaries he never lengthens a short final vowel.) Whether this license extends to the strong caesura in the fourth is doubtful: there is clarét in iii, 366, but that is all, except perit at vi, 262, which has i always in the best poets. I have not found any mediaeval poet who habitually allows lengthening in the fourth foot, although mediaeval theorists admitted it both there and in the second, quoting Virgilian precedent (see Reichling on Doctrinale, 2414, and Sprcvu- LUM, tt (1928), 360 n.).
1 Dryden notes that Claudian perpetually ‘closes his sense’ with a Golden line (9 times in 22 paragraphs in De Raptu Pros., Bk. i). Lucan also employs it to excess, and Ovid is very fond of it. It is more a rhetorical than a metrical device, connected perhaps with the late rhetorical rule that an adjective should be separated from its noun.
t ( fe lo Vi vi iv,
eon
The Bellum Troianum of Joseph of Exeter 51 False quantity is extremely rare: all cases are quoted below.
Golden lines: i, 487, 488, 489, 496, 499, 500 (6 in 14 lines, of which 5 have adjective at caesura, with its noun at the end — 3 with rhyme); iv, 172, 176, 182, 184, 185, 191, 192 (7 in 21 lines, of which 6 have rhyme); iv, 503, 505, 518, 522. Golden lines end three of four paragraphs at vi, 395, 414, 419.
Strong caesura at second and fourth foot (usually weak in third). 9 cases in i, 95-106, 5 in i, 116-121; (combined with rhyme i, 219, 225, 227, 233); 12 cases in i, 388-402, 5 in 5 lines i, 418-422. With the word iter in the 4th: i, 140, 170, 408, 521; iii, 277, 347 (425); iv, 189 (508); v, 5, 44 (160); vi, 781, 871 (except in the cases bracketed, the first foot-and-a-half consists of the adj. which qualifies iter). The form is nearly always —-vy—|——v]v-||.
Varia. excitus, i, 114; ii, 208; iii, 312; iv, 250 (cf. Brunellus, ed. Wright, p. 74, 1. 10; 123, 1. 8; 135, 1. 10 up: Specutum, 11 (1928), 362). réiectus, vi, 60 (analogy of réicit?); artériae, iii, 96; geométer, ii, 126 (cf. SpEcULUM, 111 (1928), p. 364); suspicio, ii, 192; spirius, ii, 515 (Henricus Septimellensis, iv, 153; Graecism., ix, 188; Hugo Sotovag., p. 224, 1. 12 up, ed. Wright); Phildctetes, iv, 326; Mérion for Mériones, v, 258, 278; febri- citent, vi, 82 (Architren., ix, 479, Baudri, excvi, 308); luscinia, vi, 384. Metrical license: Priamides, iv, 33, 58; Pélydamas, vi, 676.
II. GRAMMAR
Of Joseph’s grammar there is little to say: he imitates and sur- passes the bold constructions of Silver Latin, but he rarely sins through ignorance. The fewness of the irregularities collected below (of which fewer still are definitely unclassical) shows how closely he follows his models, and how vastly he surpasses his contemporaries.
A. Accidence
Noun. Constantly used adjectivally: (see VocaBuLary, pp. 75, 76, be- low) iudice causa, ‘the facts of the case determining,’ i, 375; fossor rostrum, vi, 384; uisus excubitor, iv, 181; uictor uultus, iv, 90; praeco uultus, iv, 100; uector uterus, iv, 81; euersor Asiae uultus, vi, 956; pugil = pugnaz, v, 32, vi, 492, 688; fossor ictus, vi, 561; cantatrix burus, i, 511; praeceptrix tuba, vi, 166; cornus uenatrizx, i, 509; interprete uultu, iii, 336; uindice uultu, iv, 169: uindice actu, iv, 106; indice curru, iv, 387; indice aure, iv, 236; augure plausu, iv, 238; leno oculus, iii, 244; opera nutriz, iii, 351; licito maiore,
1 Virgil rather affects this rhythm (with rhyme), e.g. Aen., vi, 299, 327, 374. 2 Cf. interprete lingua, Hor., A.P., 1. 111 (nutus interpres, iii, 242).
e e T h h
52 The Bellum Troianum of Joseph of Exeter
‘with greater license,’ v, 530; necesse (a noun), iii, 111, vi, 257; welle and posse as nouns (often in Middle Ages), e.g. iv, 420; ultra uelle nihil, nil citra posse; Salamina (nom. sing.), ii, 156; Merion = Meriones, v, 258, 278 (from Dares, 19).
Pronoun: istos = hos (as often in Middle Ages), ii, 179; sese, ‘each other,’ iv, 14; alle quidem (wrong use), v, 300; haeccine, ii, 52; v, 63, but hicne, ii, 261 (cf. 268); altera = aliena, v, 282; quisquis, ‘all’ (often in Middle Ages), v, 56 (?).
Adjective: denus = decimus, ii, 534; altus as p. part. of alo (‘plump,’ etc.), iv, 183; ili, 4, a. sanguis (“fat victim’); progressior (“more advanced’), iii, 168; fastiditior usus (“more fastidious’), i, 77: cf. vi, 777, in mensas noctis productior usus.'
Iunonior (so P= nouercatior, 27: read nouercalior), i, 268 (see SpPEcuLuUM, 111 (1928), 359); woce fauenti (abl. abs.), iii, 359; tanti = tot, vi, 104, 629; asperior = positive, iii, 280 (see SpecuLuM, 111 (1928), 355); suis liuere secundis (‘grudge his luck’). iii, 268; indigena, fem. adj., iv, 26.
Verb: depingier, 1 Arg., 5; peregrinat for deponent, i, 74, 516; transiet, future indic., vi, 34 (not uncommon in class. poetry; Anticl., 11, 304, Wright, Voigt on Y sengrimus, p. xlii); réperit (perf.), iii, 469 (SPECULUM, 111 (1928), 355), but Brunel., p. 125, 1. 14, has @ in the perf. Dietrich, Pyr. et Thisbe, |. 160 has repperiens, |. 163 réperisset.
B. Syntax
Noun: Most striking is the transference of the ablative with plenus: nata cerebro plena ( = pleno) Iouis, ii, 429 f.; cf. plenum galea caput = galea plena capite, v, 253.
Genitive: fraudis hebes, iv, 120; caeca freti, iv, 298; dubius aquae, i, 83 (so P); fas precandi (‘chance of prayer’), iii, 248.
Dative: Iunoni celebris (‘thronging to worship’), iii, 207 (?); urbi medi- antibus aruis (‘fields coming between [it and] the city’), i, 505.
Accusative: exire lares, vi, 782.
Vocative: Sic Helenam uisure ueni, v, 324; cf. i, 286. Joseph affects the vocative of participles in -us.
Ablative: agent (without ab) laudor censore caduco, ii, 244; principe iussus, v, 453; habetur Pallade (= Palladio), i, 536.
Verb: Most characteristic is his use of the fut. participle, as in Silver Latin (6 in St., Theb., vii, 776-794), e.g. i, 158, 164, 169, 521, 532; iii, 434, 437, 439; iv, 517, 519; v, 225, 226, 229; non sensurus (‘unfeeling’), v, 198. Connected with this is his use of the gerundive as a fut. pass. ppl., e.g. frangendus (‘doomed to be broken’), i, 120 (“breakable,’ i, 267); ef. iii, 343;
1 For other cases of participles in the comparative (a favorite device) see pp. 75, 76 below. 2 See p. 69 below.
t ( u fi P P in (c ci A fo CO uor
The Bellum Troianum of Joseph of Exeter 53
vi, 712; strange is ii, 175, rapienda minatur ( = se raptum iri). Character- istic also is the extended use of the prolative infinitive: e.g. with negat (‘refuse’), ii, 240, iii, 118, v, 22, 322, vi, 260: only quoted in lexica from Silius, but see Claud., Ruf., ii, 223; cf. Voigt on Ysengrim., p. 1; sociare superbit (‘scorns to’), v, 180, vi, 490; bellare proterua, vi, 503; negat ire (‘forbids to go’), iv, 450; impatiens ( = non patiens) abesse, v, 446. This is combined with a perfect infinitive for present in ii, 28, exorant habuisse; iv, 392, uolo latuisse (with iubeo, i, 431): Cf. Voigt, p. xlvi. monstrorum domitor reminiscere, i, 284 (‘remember that you are . . . ’); ef. rapienda minatur, above.
Conditions: si haesit (for fut.-pf.), ii, 246; si with imp. subj. for ppf., i, 491 f. (very common in Middle Ages).
profando = profata (very common in Middle Ages), ii, 315.
ausa (‘by an act of daring’) ii, 277 (see SpecuLuM, 111 (1928), 372).
moderantior = -atior, ii, 422, iii, 235, v, 216; either imitation of a late usage (Stolz-Schmalz, Lat. Gram., 4th ed., Munich, 1910, pp. 236, 299, 481) or moderantior may be comparative of the Lucretian moderanter).
Ellipsis: (rogauit) ne quicquam temere, iii, 194.
mon... que = neque, ii, 527 (but read -uwe?); -que tamen = tamen, vi, 111; ut quid? (‘why?’), i, 6.
retro iter (“way back’), v, 468.
Ill.
While normal in verse and grammar, and closer to classical usage than other mediaeval poets, in style Joseph casts aside all restraint. One would have thought that the ever-increasing ingenuity in the use of every conceivable rhetorical device and contortion of speech, from Virgil onwards, had reached its culmination in Statius and Persius; but a study of Joseph will show that there were still further possibilities. It is true that he avoids the mere childishness taught in the schools, of which Matthew of Vend6éme is perhaps the best (or worst) exponent, and Architrenius the most thoroughgoing dis- ciple — see, for instance, the description of a maiden at the end of Architrenius, Book i. Here ingenuity of expression is merely a cloak for exiguity of thought;! but in Joseph the case is reversed. Thoughts come crowding on each other in such profusion that language is left
1 I give one case from Planct. Nat., p. 482 (Wright): dum Charybdis ingluuiosae hyatus uoraginosos subterfugiendo deuitant, in Scyllae malignantis abysso inopinata periclitatione mis- erabiliter naufragantur.
| | i J [ t ii i } | | {
54 The Bellum Troianum of Joseph of Exeter
panting behind. To compare small things with great, we are re- minded of Bagehot’s dictum that if Shakespeare had tried to write an epic he would have died of plethora of thought. But here it is the ingenuity of a rhetorician which takes the place of the imagina- tion of a poet: I doubt if there is a single line of poetry in the highest sense in the whole six books. And over and above the ingenuity of thought, comes an almost incredible ingenuity of expression. Let me quote an extreme example:
Sic fatur, diroque animam depascitur aestu; Has tamen, has cordis rugas, haec praelia mentis Frontis oliua domat, faciesque dolosa sophistam
Mendicat uultum, et blandos peregrinat in usus. (i, 286-239.)
All he means is that the man conceals his anger, but he uses seven metaphors, most of them unparalleled in classical poetry (those of 237 smack of the schools: cf. examples from Poetria Noua, Specu- LUM, 1 (1928), 358: an even worse case is ii, 124 ff. (Hercules and
Antaeus): Pestem Libycam librauit in auras, Edocuitque hostem nocituris artibus astra, Qui quondam geometer erat...
i.e. raising him aloft, he taught by painful experience astronomy to one who had before been an earth-measurer. More typical is the opening Book iii: a careful study of this will reveal many of the secrets of Joseph’s workshop (I do not comment on every point):
Soluuntur uario consulta silentia plausu:
Idalium paeana canunt, plebs, aula, senatus:
Hic fatis uouet, ille deis, una omnibus, una
Voce animoque frequens plaudit Venus; altus ubique 5 Sanguis obit, sacris ultro cessantia cedunt
Arua, coronato luget spoliata marito
Inachis, et raptos gemebunda reposcit alumnos.
Tllustres superis epulas molita potestas
Sacrificis bibulos lictoribus instruit ignes,
10 Totaque thuricremis Panchaia spirat in aris, Venales emptura deos: at pauper acerra, Principe digna Deo, nostro placitura tonanti, Mente litat pura, uotis exorat honestis.
Celsior explicitas rapturus in ardua flammas
15 Regius exstruitur congestis floribus ignis, Perstringitque aciem: procul his expirat ab aris
l t a c st tl D pl D bl cil
1e
The Bellum Troianum of Joseph of Exeter 55
Hostia dira, cruor; Veneri quod dulce propinat Ductor Aristaeos latices, Meliboea fluenta, Icarios haustus, fracti Phoenicis odores.
In 1, consulta silentia = ‘the silence of the council.’ In 2, Idalium paeana is no doubt an intentional oxymoron. In 4 f. frequens plaudit Venus = ‘the name Venus resounds from many a throat’ (plaudit = sonat, as ii, 86); altus, p. part. of alo (as in Virgil); sanguis obit, cf. 1. 16 cruor expirat. Sacris . ..seems to mean ‘yielding place to sacrifices, the fields are idle,’ i.e. ‘the plough oxen are sacrificed.’ In 6-7 note rhyme of noun and adjective — the effect in both is that of a Golden line: 9-10 are Golden lines, and 8 has the same effect. Inachis =Io = a cow. 10 is an ingenious refinement on Virgil’s totaque thuriferis Panchaia diues arenis. 11-13 each begin with adjective agreeing with noun at caesura (or vice versa) and end with another noun-adj. pair. 14 again has Leonine rhyme. 15 is a Golden line. 16 ff. are difficult. The meaning apparently is (I have altered the punctuation of 17) ‘far from these altars [i.e. not here] is blood sacrificed: the King offers what is pleasing to Venus, viz. honey, milk [Joseph learnt from a commentary on the Eclogues the meaning of Meli- boeus], wine, and incense [fracti =‘aged’].’ The whole is in imita- tion of Statian circumlocution; see Dresemius, ad loc. He not only avoids saying ‘milk,’ ‘wine,’ etc. but even avoids using the plain circumlocution with liguorem. A bloodless sacrifice is clearly de- scribed (cf. 29 f., 46) yet 23 seems to say a dove is sacrificed: does the line mean something else ?
It is instructive to compare Joseph with his immediate source, Dares, in the descriptions in Book iv (see Root, art. cit., where Root proves that Chaucer deliberately imitates Joseph).
Perhaps the most striking is the characterization of Neoptolemus. Dares has Neoptolemum magnum, uirosum (‘manly’), stomachosum, blaesum (like Hector), uultu bonum, aduncum, oculis rotundis, super- ciliosum. See what Joseph makes of it!
Ore uirum fassus, indignarique minaces
Suspendens ceruice genas, liuore superbo
Obliquat socios Pyrrhus, scribuntque ferocem
Ardua membra, teres oculus, proiectior aluus, Texendis uox aegra sonis, teneroque laborans Eloquio, lingua blaesos frustratur hiatus. (iv, 136 ff.)
: { | ; n i d te A j
56 The Bellum Troianum of Joseph of Exeter
In 1. 136, uirum = manhood, indignari depends on minaces (both typical usages, see examples of grammar and style); in 2 ff. the use of the verbs is characteristic (obliquat, ‘looks askance at’), so is protectior and the use of texendis, and the adjective aegra: in 4 teres = Dares’ rotundus: Joseph must have been thinking of Horace’s teres atque rotundus.! Lingua looks like nom. sing.: if so, this is the only case I have noticed of lengthening a short final vowel at the caesura.
It is safer to make it ablative.
In spite of its length, I cannot forbear quoting the description of Helen. Mediaeval poets vied with each other in full-length descrip- tions of female beauty, as prescribed by the textbooks, so we may assume this will represent Joseph’s art in its highest elaboration:
Inclyta cognatos aequat Spartana Lacones (Golden line) Vultu, crine, genis: sociae cunabala stirpis Par oris testatur honos; at plenius haurit Sidereum | Ledea Iovem ||, totosque par artus 5 Lactea materni spirant mendacia cygni. (Golden line)
Natiuum | frons fundit ebur ||, caput exerit aurum Aequatis | crinale comis ||, gena candida byssum (note symmetry) Aequat, dextra niuem, dens lilia, colla ligustrum. (ab, ab, ab, ab) Auris inexpleto crispata uolumine, uisus
10 Excubitor, nasusque uagi uenator odoris, Poscunt alternum titulo certante triumphum. (Golden line) Productum | modice mentum || candescit, et alta Lenius ut pressis | insidant basia labris ||, (quasi-Golden line) Parcius in roseum | consurgunt ora tumorem ||. (do.)
15 Fundit apex humeros, pressum tegit ubera pectus, (note symmetry) Tlia succingit leuitas, tractusque lacertos: Pes breuis incessu | terram libante supino || (quasi-Golden line) Lasciuum | suspendit iter || crurumque decora Mobilitas comto libramine ponderat artus. (quasi-Golden line)
20 Sola superciliis labes interflua raris Audaci macula tenues discriminat arcus (edd. artus) (Golden line)
I have indicated the extreme artificiality of the metre: I add a para- phrase:
She resembles Castor and Pollux: the origin of a common descent is indicated by equal beauty, but she has imbibed more of Jove, and her milk-white skin breathes the deceit of the swan which tricked her mother. Her forehead, hair, and so forth, are like ivory, gold, and so on. The incom-
1 Joseph seems to have thought teres = rotundus: cf. iv, 88, gena tereti (= Dares’ ore rotundo); iv, 150, teretes artus (= Dares’ rotundus).
Ss
&
ara-
at is
her ther. com-
ps ore
The Bellum Troianum of Joseph of Exeter 57
plete circle of her ear, her eyes — the body’s watchmen — the nose track- ing the fleeting odors, alternately demand the prize, with competing claims. Her white chin slightly protrudes, and that rich (alta) kisses may rest more gently on her lips when pressed, they delicately rise in a rosy swelling. Her shoulders spread out from her head, a subdued bosom re- strains (or tops) her breasts, her flanks are light and slender (succintus), her arms long: her foot is small and she skims the ground with the sole of it (supino?), poised with dancing gait: her light legs keep her body delicately balanced: a single mole between her eyebrows divides them into two slender arches by a bold splash.
I think this will be enough: but after describing her exterior, Joseph gives us a physio-psychological description of her interior — her lungs are, of course, for talking (modulamina linguae limat pulmo loquax), the spleen causes laughter (sum petulanti splene cachinno, Persius, i, 12, apparently misunderstood), bile anger, and the liver love — and there’s the rub; ‘sic Helenam totam pars unica mergit,’ her too active liver spoils all! This seems to have been part of the contemporary doctrine of physiology: we find in the Grecismus: ‘cor sapit et pulmo loquitur, fel commouet iram, splen ridere facit, cogit amare iecur,’ xix, 106 f.! It would be a mistake to laugh at all this: it was evidently admired in the Middle Ages, and Chaucer was not above imitating it — not only in the Parlement of Foules, vv. 176-182 (cf. Root, art. cit.), but in his ripest work. Joseph simply beats the rhetoricians at their own game.
Fairness demands that we should now quote some more favorable passages. For vigorous rhetoric Joseph equals any of his Silver-Latin models, e.g. i, 59 ff.
Cui iam satis est quod regna, quod urbes,
Ipsaque quod pateant rapiendis Tartara gazis?
Itur in ignotos fluctus, ultroque procellis
Insultare iuuat, et soli uiuere fato. We should have to search a long time in mediaeval poetry to equal that. Or this:
O superi, quo thura meant? O fata, quis error
Imperat? O hominum casus! Paucosne reseruat
Ut multos Fortuna premat? Iam sentio, traxi Inuito mea regna deo . . . Voluine sub Arctos
1 The source seems to be Claud., 1V Cons. Hon., ll. 241 ff.: cf. Isid., xi, 1, 125.
| es ly a. of p- ay
58 The Bellum Troianum of Joseph of Exeter
Proscripsisse uirum? Regnat. Fregisse? Superbit. Extinxisse? Viget. Titulos minuisse? Triumphat. Ergo deos dita (edd. dico) ingratos: da thura tonanti; Cum dederis, sperata neget, donisque retentis,
Rideat elusum: melius Fortuna pianda est,
Quae pacem, quae fulmen habet, quae prima tyrannis Imperium partita tribus, concessit, abegit,
Mutato lusit caelo, rursusque iubebit
Quos fecit reges mecum migrare sub umbras,
Et Iove proscripto Saturni leniet iras. (i, 216 ff.)
(The last line rather spoils the climax, and it is followed by the shocking lines quoted above, p. 54, sic fatur, etc.) Another effective passage, still in Book i, begins (vv. 436 ff.):
Parcite, uictores, Danai, compescite dextras: Labe pari peccat, pariter crudelis uterque, Que cunctis parcit et nulli, nobilis ira est Citra iram punire reos.
Or, take the following: miratur Achaea iuuentus, Duxque iterum: superisne manus et tela negamus, Inachidae? uocat ecce deus: sol laetus anhelis Instat equis, Neptunus aquas ac peruia mollit Caerula: credo equidem uires (ui res?) a gentibus istis, Laomedonteo poscunt ex hoste triumphum. Sed moror. Exclamant Danai, totoque resultat Vox animosa freto. (v, 106 ff.)
Heu, heu, quam tenui nutant mortalia filo!
Nil homini fixum: fortunae munera blandae
Insidias, non dona reor, semperque timebis,
Sirenum turbae similes, sub sole sereno
Nuben, sub risu lacrimas, sub melle uenenum.
Si tibi res, fallit casus; si forma, senectus;
Si uires, morbus; si nomen grande, litura
Postera, et in nullis fati constantia donis. (v, 511 ff.)
tandemque dierum Plenus obit, felix senio laetusque iuuenta, Si stabili fortuna fide, si fixa maneret: ! Post Zephyros plus laedit hiems, post gaudia luctus; Unde nihil melius quam nil habuisse secundum?. (vi, 801 ff.)
1 Cf. ii, 24, felix coniugio, felix natalibus aruis: si superi, si fata sinant, si stare beatis permissum.
2 secundum ‘prosperity,’ but there may be a pun on the other meaning. The sense is exactly Dante’s nessun maggior dolore, Inf. v, 121, rather spoilt by Joseph’s cleverness: Jusserand quotes the last two lines only to condemn them, but how superior are they to Lucan, viii, 29 ff., their source!
he
ve
ness: ry to
The Bellum Troianum of Joseph of Exeter 59
Andromache’s dream (modelled partly on Aeneas’ vision of Hector in Aeneid, ii) achieves true pathos, but is too long to quote here (vv. 423 ff.). Like his models, Joseph excels in sententiae, e.g.:
tempora certe uirtutem non prima negant, non ultima donant. (i, 21) premit ultio noxas tunc gravius cum tarda uenit. (i, 38)
litigiis audax et citra proelia fortis. (i, 317)
pronius in facinus, tardis ultoribus, itur. (i, 378)
sibi quisque uideri dux aliis, milesque sibi. (i, 394)
nam summe miser est cui nec sperare relictum est. (i, 474)
dignetur nemo priorem, erubeat sibi quisque parem. (v, 78)
cur uixi? heu, miseris longo nil tristius aeuo. (vi, 826)
The above specimens suffice to give a good general idea of Joseph’s merits and defects, his luxuriance and over-ingenuity, his astonishing mastery of every rhetorical device of classical poetry, and his frequent bursts of vigorous and fiery eloquence. It is the luxuriance of the young Shakespeare of Lucrece and Venus and Adonis, or of the young Pindar who ‘sowed not with the hand but with the whole sack,’ and we may well suppose that when Joseph mellowed down he might have produced something of enduring merit. Unfortunately his other work (notably an Antiocheis on the Third Crusade, promised i, 47 ff.) is not extant, and we shall probably never know whether he fulfilled the promise of this early work. As it is, Joseph does not attempt to palliate or deny the ‘atrocious crime of being a young man,’ rather he glories in it (i, 19 ff.), and he would no doubt have answered his critics in the words of the young Accius to the veteran Pacuvius, who found his grandiose style was marred by immaturity and harshness: ‘I admit the justice of your criticism and would not have it otherwise: time will mend that.’
Let us now take a more detailed survey of Joseph’s style. Though it is hardly possible to illustrate it exhaustively in a few pages (even were that desirable), one can perhaps give some idea of the bewilder- ing variety of expedients Joseph can command. The conduct of the story was the last thing he troubled about; in fact he seems, like Lucan, mainly concerned to get away from the story and concentrate on speeches, descriptions, rhetorical outbursts and moral and theo-
| Hi
se is
60 The Bellum Troianum of Joseph of Exeter
logical disquisitions. For instance, being bound by the rules of the game to introduce gods and goddesses, he either makes them a peg on which to hang satirical disgressions, or, as in the Judgment of Paris, puts in their own mouths speeches which lay bare the crudi- ties and puerilities of heathen mythology. Like his model Statius (and his model Antimachus) he take up half his poem before he gets to the story proper. A synopsis of part of Book i will illustrate the ‘tricks of the trade.’
Proem and dedication (1-51). The invention of ships due to greed; greed scours the earth and flouts the elements (52-62). The Argonauts were the first sailors (63-69). The first attempt was on a modest scale (70-82). The crew hesitate (83-86). Why ask for trouble and invent new deaths? (87-90). Sea do your worst! (90-93). But the sea-gods welcome new votaries (94-108). The Argonauts land at Sigeum (2 lines!). Uproar at Troy (111-115). Why fight the favorites of the gods, O Trojans, doomed to destruction? (116-128). The fight begins (129-135). Nestor advises patience; speech (138-149). The Greeks sail on to Colchis (2 lines). De- scription of harbor and landing (151-173). Why describe the plowing, the dragon, etc.? (174-178). What a wicked conspiracy! (179-190). They return (191-194). What rejoicings! (195-205). But Pelias is angry (206- 215). His speech: ‘What are you about, O Gods?’ (216-235). His dissi- mulation (236-239). Victors’ feast (240-258). [Satire on luxury 241—248.] Hercules cannot rejoice (259-262). His self-reproach (soliloquy) (263-290). Compared to bull (291-297). Remarks on Fama (298-307).
An analysis of the above gives us 59 lines of satire, 44 of descrip- tion, 120 of rhetoric, 51 of preface — and 19 of narrative!
An extreme case is the Judgment of Paris in Book ii, where after some 400 lines of description and speeches, Joseph abruptly termi- nates the book, and forgets to give the judgment at all!
I shall now attempt to classify in detail some of the most note- worthy features of Joseph’s style.
A. Use of the Noun
fur oculi somnus, ii, 238; metu praedone animi, v, 50; custos animt timor, v, 469.
Periphrasis for person: bellum consumite (= bellantes), i, 368; alter cum Polluce error (i.e. Castor), ii, 176 (cf. gratus parentibus error, Aen., x, 392); funera exilii tui (‘of those you exiled’), ii, 20; occursus omnes rogat (‘asks all she meets’), v, 448; armant fugam (= fugientes), vi, 152.
cc a he 2 in un Pi vi al; G: vi, we Vi,
‘mt
um 2); sks
The Bellum Troianum of Joseph of Exeter 61
Person for quality: eruo regem (‘kingship’), ii, 42 (cf. exuo fratrem; Ant., p. 299, 1. 4 up); dedecoro regem, vi, 799; regem profatur, iv, 101; reginae meminit, ii, 316; similem animis indutus Achillem (a pupil of Chiron become a second Achilles), v, 242; weniunt cum principe (‘kingship’) pestes v, 292; liquidus Iupiter (= Neptune), i, 286; Phoebus uterque (sun and moon?), vi, 664 (Dresemius’ ‘rising and setting sun’ does not give the required sense); detersa meretur Iuno Iouem (‘the cleared-up atmosphere wins back a clear sky’; Juno = aer inferior, = i, 243), iv, 311; Atropon auzit (‘added to the ways of death’), i, 187; inuersa est uetus alea (‘old fortune changes’), iii, 44; pulchra genarum rapta dies (the eye), v, 369.
B. Use of the Verb
With characteristic use of in: animum procudit (extendit, =) in ausus, i, 290: for this use of in cf. iv, 65, 125, 144, 156-157 (3 cases), 159, 165, 288, 359, 458, etc.; in uoces dissuta breues (‘voice broken by sobs’), vi, 816.
medicati uulnera muri respirant, i, 477; dura angere (‘crush misfortune’ (by fulfilled wish)), iii, 342; armati aéris ira, iii, 438; ramos loquaces ore premit (of seer, dadvqdayos), iii, 364; funera librat (‘kills equal numbers’), v, 355; mediumque coronat armata statione wirum, vi, 426 (‘surrounds’); ultricem cogit plebescere dextram, vi, 348 (from Stat. Theb., xii, 738; cetera plebero desaeuit uulnere uirtus); fregerat (v. 1. fugerat) undas Phoebus (“broke into flashes (ripples) ’), ii, 269 f.; saeuos diffibulat urbem murus in egressus, v, 131; pennificat formido gradus, v, 203 (from Virgil’s pedibus timor addidit alas, but is pennificat right in this construction?).
C. Use of the Adjective
Cirrhaea iuuenta = young laurel (on Mt. Ida!), iii, 218 (st wera lectio) ; consulta silentia (‘silence of the council’), iii, 1; ardens somnus (‘dream of a torch’), iii, 157; cognata oracula (‘native (local) oracle’), iii, 302; sociae ruinae (‘death of friends’), v, 262; mellita pericula (‘fight (of bees) for honey’ (?) ), vi, 19; spondeo stabilit animos pede, vi, 87 (from Hor., A.P., 256; spondeos stabiles); reditus Ledaeos (‘of Helen’), vi, 673; Herculeis imbuta uadis (ref. to the poison of the Lernean hydra), vi, 233 (cf. Herculea unda, iii, 289; Hectoreum femur, v, 263; Hectoreos manes, vi, 75; inscitia Plisthenia, v, 286); campestres iterare manus (‘renew fight on the plain’), vi, 565; fraxineo peplo (‘banner on ash pole’), vi, 583; fulminei Tyrias pars alitis alas incestat Ganymede suo (‘defile with a picture of their native Ganymede the purple wings of the lightning-bearing eagle’ (of a banner)), vi, 586; Carpathias linguis aequare figuras (‘equal the numbers of the waves’), vi, 764; somnus uigil (‘awak« ving from sleep’), vi, 773 (but in vi, 778 uigilata quies = ‘keeping awake’).
le | of i- 18 : | 1e d; re ), s? at od es he ey | si- 8.] p- er ni- te-
62 The Bellum Troianum of Joseph of Exeter
D. Other Cases of Involved Diction
segni morantes crescentesque cibos gustu ieiuna fatigat, ii, 153, seems to mean ‘she tires the lingering food by the slowness of her eating’; minus uno in natum producte gradu, ii, 161, ‘advanced to sonship by one step less,’ i.e. he would have been his son if he had not been his grandson! (?); frequens plaudit Venus, iii, 4 (cf. wlulatum ‘vale,’ iv, 494); fessa reclinat flamma apices, iii, 48; indutus Nessum, iii, 91 (cf. indutus ignes, iii, 145); praesentius audet (‘has readier daring’), iii, 274; armatos in pacem reddidit annos (‘gave me years of peace after war’), iv, 435; prima ducis facies nascenti uagit in aere (‘the bronze depicts his earliest childhood’), iv, 462 (cf. accingens pictura uirum ‘represents him in armor,’ iv, 471); woluntas omnis in effectum causas habet (‘will always gives motives for action’), v, 382; decimus rotabat speratum Titana labor (‘the tenth year of war brought the hoped-for day’), vi, 661 (cf. denus tumebat mensis = tenth month of pregnancy, ii, 534).
We may compare the way he paraphrases Dares, e.g. in the descriptions of Bk. iv. uisus acies obliqua cadentis in geminas spargit partes (Dares’ strabo), iv, 54; gratior aequato Menelaum forma uenustat fulta statu (Dares’ mediocri corpore), iv, 104; wirusque comae geniale (natural?) rubentis mens domat (Dares’ rufus: but a mediaeval writer has to explain how a red-haired person can have any good qualities),' iv, 105; in pectus laterum laxatur culmen (Dares’ pectorosus), iv, 110; wox en eris (teneris?) discincta iocis (Dares’ iucundus: Joseph falls into the usual confusion with iocundus), iv, 116; after describing Briseis’ hair, wmbraeque minoris delicias oculus tunctos suspendit in arcus (Dares’ iunctis superciliis: the ‘delights of lesser shade’ are the eyebrows!), iv, 158 (cf. umbra supercilii, Claud., Nupt. Hon. et Mar., 268).
Joseph is fond of catalogues; of course no epic is complete without a catalogue of trees. Virgil (Aen., vi, 179 ff.) imitated Ennius, and was in turn imitated by Ovid (Met., x, 86 ff.), Seneca (Oed., 532 ff.), Lucan (iii, 440 ff.), Statius (Theb., vi, 91 ff.), Claudian (De Rapt. Pros., ii, 105 ff.): Joseph starts a new line (i, 505 ff.), though he knew all his predecessors, and was in turn imitated by Chaucer (Parl. Foules, 176 ff.; cf. Root, art. cit., p. 18) and Spenser (F.Q., 1. i, 8 f.): each tree has its appropriate epithet, which is so ingenious that all have to be explained by the commentator in the Paris MS., e.g.
1 Cf. suspectus crinis, iv, 88. I may perhaps be allowed to refer to my semi-popular article on the subject in the English Review (June, 1927): note Anticlaud., rv, ii, 16 f.
fc le m pe In lig wl but Yse In tl she | scho The |
The Bellum Troianum of Joseph of Exeter 63
uaga pinus, because ships are made of pine (Chaucer’s ‘sayling firr’; cf. Spenser, loc. cit.; cantatrix buxrus, because flutes are made of box- wood (Chaucer’s ‘boxtree piper’).! A mediaeval trait appears in lines of the form nil refert quis, ubi, quando sed qualiter et quid (vi, 33); in quam, quid, quare (ii, 492); cur, quo, unde (iii, 201). Whatever Joseph does, he does with all his might; if Virgil’s prophetess is mad (Aen., vi, 241 ff.), his must be madder (iii, 366):
claret interprete uultu Extorti uindicta dei, furiosa potestas Imbelles uexat artus, ceruice rotata, Crine umbrante humeros, oculis spargentibus ignem Et facie (MS. facula) prodente fidem, nunc liuida uitrum, Nunc flammata facem superat, nunc pallida buxum. Tali se mediis procerum tot milibus infert Ore ferens, tremulosque gradus uix ebria figit.?
(The first words mean: ‘the revenge of the god for having been forced is manifest, as interpreted by her appearance . . .’)
Joseph is revolted by no violence: in v, 253 f., a decapitated head continues speaking;* in v, 416, the hero kills seven at a blow; in vi, 228, the spear enters the back of the head and comes out taking the left eye with it; in v, 350, Aeneas traiicit, et fixae ‘moriere’ immur- murat auri; in vi, 158 ff., Thetis, for pity of Ajax, altius attolens pontum . . . wenienti occurrit Phoebo — and so makes the sun set. In ii, 430, the virgin Minerva leaves incomplete the word her chaste lips refuse to utter; cumque exule mentu . . . erubeo ulteriora loqui.*
Like other mediaeval poets Joseph pushes oxymoron to a stag where it borders on the absurd; an extreme case is ii, 418 f. (of love)
1 On this tradition, see the notes in Classical Weekly, xx11 (1929), 66, 184.
2 Yet even this is better than Lucan’s prolix ineptitude, Phar., v, 169-192.
3 Cf. Ennius, Ann., 472 f., 519 f.; Virgil, Aen., x, 395; Ovid, Met., vi, 556; Statius, Th., xi, 56 ff.; Silius, xv, 470; Zola, Débdcle, ii, 7.
* The phrase exul mentula itself, referring to the mutilation of Cronos, is typical of Joseph; but he was not the only mediaeval poet who indulged in this novel form of aposiopesis. In Ysengrimus, 1, 928, the speaker is interrupted by a howling mob at the words tu au(-tem). In the Speculum Stultorum the dying caif has not the strength to name her friend, Brunetta — she can only gasp out Bruna, uale! It almost looks as if this were a device taught in the schools, but I can find no trace of it. In later times it has been relegated to burlesque, e.g. The Critic, 111, 1 (see my note in Times Lit. Supp., May 26, 1927), where Gustavus Vasa is paro- died: a notable modern use is at the end of Verdi’s Otello (last page), ‘un bacio, un ba . . .’
| to | us | at ; lit es 62 as rar ith ns es’ es’ ons red i tur cis 3), lus ser put nd 4 pt. ew arl. f.): all e.g.
64 The Bellum Troianum of Joseph of Exeter
uirus dulce, piae Syrtes, amentia (‘rage’?) supplex, molle malum, morbus hilaris; this, though an extreme case for Joseph, is mild com- pared with some mediaeval examples.
Likewise in his use of paronomasia,' Joseph is a child of his age; although this is one of the earliest ornaments of Latin style (for instance in Plautus and Ennius), and no doubt native to the genius of the language, it was very sparingly used in later times. Among the most striking examples are: i, 19, mento canescant alii, nos mente; i, 72, e ramis, remos; i, 351, colles colit; ii, 96, arteriae non artus; ii, 131, wector eras et uictor (MSS uector) erts; ii, 254, Juno calls Minerva, by a pretended slip of the tongue, dira for diua; ii, 286, una sed unica; ii, 288, absque pari parit; ii, 413, wenatrix hominum uernet Venus, and later (420) exitium uenale Venus; iii, 48, solatam solare domum (‘console a desolate house’); iii, 108, audit et audet, cf. iii, 240, ardet et audet; iv, 382 and vi, 361, fugit, fugat; vi, 847, thalami tumulique; vi, 961, uiue liber, liberque uige. Akin to these is the passage (ii, 271 ff.) where Juno suggests that Pallas was so called quia pallet, and Mauortia because mares uorat (an ancient deriva-
tion?); while in her reply Pallas points out that the true derivation ( is from polleo. P Joseph’s use of alliteration is moderate: the most striking cases c are: i, 284 (5 p’s); i, 294 (f and p); i, 409 ff. (12 p’s); i, 453 (p); W i, 477 f. (m); ii, 413 (v); ii, 490 (f); iv, 50 f. (c); iv, 337 (p); iv, 388 (f); fo v, 32 (p); v, 41 (v); v, 486 ff. (p); vi, 162 ff. (c); vi, 452 (c); vi, 663 as (5 m’s). hi But in spite of all his ingenuity Joseph cannot always avoid la monotony; this is perhaps chiefly due to his metre, though there are his other contributory factors. First comes his excessive use of sym- fat metrical arrangements of words, particularly in triple pairs, either (a1 in the form aaa, bbb, or ab, ab, ab. In fact Joseph’s great defect is BH su; that he can hardly think of a noun without an adjective, a trick rot which when twice or thrice symmetrically repeated becomes very gen wearisome. Attention to Joseph’s love of symmetry sometimes @ (m helps to a better understanding of the text, e.g. ii, 236 ff. In the Hove the
Delphin edition this makes nonsense: we should read and punctuate 1 See M. B. Ogle, Specutum, 1 (1926), 170 ff.
i mot
The Bellum Troianum of Joseph of Exeter
Si gena ( = facies), si generis auctor, si denique mores Quaerendi, in curis regnat pudor, a Ioue sanguis,
Censori facies oculo patet: haec mihi (edd. haeccine) forma, Hoc genus, hic mentis cultus.
There is a triple group of three, abe, cha, abe. Compare ii, 208 ff. (here again we can improve the punctuation):
Ocyus exciti, qui casse uel ore uel aure, Fraude, sono, sensu, ludentes proelia casse Fallunt, ore cient, uestigant aure.
And vi, 154 f.: Incendunt, sternunt, perimunt, face carbasa, contis Arces, ense uiros.
An ingenious case is v, 67 f.:
quos obuia Scylla, secuta Tellus, et socio fregit Thetis hospita fluctu:
ie. ‘Seylla before, Tellus behind, Thetis accompanying’ (this last expressed by both socio and hospita).
Then comes his excessive love of various tricks of style — strings of words, use of the future participle, ablative absolute, adjectival use of noun in apposition, over-fondness for certain words and types of compound. For instance in the 14} lines 388-402, eleven have a weak caesura (or none) in the third foot, strong in the second and fourth. In ll. 385-410 there are six ablative absolutes: the word asper (asperrimus) comes in Il. 392, 397, 401; in ]. 404 we get one of his favorite compounds in de —decaluo (which occurs in the Vulgate); iam copia muri (1. 398) is echoed by tam copia regni in |. 439. Among his most overworked words are the Statian words castigo (‘check’), fatigo, fatisco, furatur ( furatus, furtum), incesto, iugalis (‘steed’), libo (and compounds), mentior (mentitus), obliquo, roro (roratus), sidereus, superbio, titulus (‘glory’), werno. From Lucan he adopts larus (laxo), roto, spargo, uaco, uotum. Other favorite words are arduus, fastidio, geminus, nego with prolative inf., nuto, peregrino (met.), suspendo (met.), ausus-iis, and ausum used asa noun. But of all, the most overworked is libro; Statius uses it 13 times in 14 books (excluding the Siluae, which Joseph did not know): Joseph in six books has it more than 20 times, often in the most extraordinary uses (also
65 |
| ius ng us; alls | 86, um 47, e is led va- | ion (Pp); (f); 663 | roid ‘| are ym ther is | rick | very | mes the | uate
66 The Bellum Troianum of Joseph of Exeter
libramen twice, and libra). It is most often used of the poise of the body, e.g. stabiles libratur in artus (iv, 74), gressum libare (iii, 225, 330), sometimes simply of poising a weapon, sometimes in a meta- phorical sense, rerum librare uices (ili, 135), consilit librata fides (iv, 129)— see VocaBuLary, pp. 72 ff. below.
This brings us to the question of Joseph’s borrowings. His famili- arity with such Latin poetry as was accessible in his day is astound- ing; unlike most mediaeval writers, he does not clumsily imitate a limited number of poets studied for the purpose of pilfering: he has mastered the whole corpus. Yet, unlike his contemporaries, he does not as a rule borrow slavishly: what he takes, he has so thoroughly assimilated as to stamp it with his own individuality; we do not get the impression of a cento, although in a sense he owes more to the classics than any of them, for his style is, with all its eccentricities, essentially classical. But he imitates as an equal: what Statius and Claudian are to Virgil and Ovid, that — and neither more nor less — is Joseph to them. His metre, grammar, and vocabulary are essen- tially Classical; he does but carry further the process of elabor- ation which marks the whole succession of Latin poetry.
Accordingly it is difficult to apportion his indebtedness. Perhaps he comes nearest to Statius in general treatment; in satirical passages he owes much to Claudian; on moral themes he often comes nearest to Juvenal (who, by the way, often recalls Joseph’s style; e.g. genua incerare deorum, x, 55; unum ciuem donare Sibyllae, iii, 3, are just the sort of phrase Joseph loves); in his inexhaustible ingenuity he was an apt pupil of Ovid; in battle-pieces he chiefly uses Virgil and Statius. But his borrowings are usually too subtly interwoven with his own texture to be readily detached. I have collected a number of parallels — with the help of Dresemius; had my knowledge of later Latin poetry been more thorough, I should no doubt have found many others. I append a list of the most striking:
Book I
Iliadum lacrimas, |. 1 = Juv., x, 261; bis adactam cladibus urbem, |. 2: b. a. legibus Istrum, St., Theb., i, 19; wert sacra fides, 1.7 = Claud., Bell. Get., 1. 553; mens conscia ueri, 1. 28: cf. m.c. recti, Aen., i, 604; unde ruat
S pre...
Pa
pen
Eu
‘na
Bell.
The Bellum Troianum of Joseph of Exeter 67
tabulata struit, i, 38: cf. t. unde altior esset casus, Juv., x, 106 (cf. M. R. P. McGuire on Ambr., Nabuth., 4; Claud., Ruf., i, 22 f.); felix iniuria uoti, |. 41 = Claud., 1v Cons. Hon., |. 618; deum genitore metu mens caeca creauit, |. 99: cf. primus in orbe deos fecit timor, St., Theb., x, 384 (= Petron., Epigr., 27.1); omni homini commune solum, i, 123: cf. o. h. natale s., St., Theb., viii, 320 (omne s. forti patria est, Ov., Fast., i, 493); prudentia, turbae rara comes, |. 134: cf. fors, ingentibus ausis r. c., St., Theb., x, 384; procul o procul, ite profani, 1. 181: p. o. p. este p., Aen., vi, 258; paucosne reseruat ut multos Fortuna premat, |. 217: cf. quicquid in altum Fortuna tulit, ruitura leuat, Sen., Agam., 100; Fortuna multis parcere in poenam solet, Pub. Syr.; and nearer still, Caes., B.G., i, 14,4. Cf. other passages in Dresemius’ note: sponte deorum, |. 222 = St., Theb., v, 136; uenantes uenata uiros, |. 247: cf. uenatrixque metu uenantum territa fugit, Ov., Met., ii, 492; 1 nunc et . . .i, 270, v, 323: Mayor on Juv., x, 310. Simile of bull, ii, 291 ff.: cf. Lucan, ii, 601 (from Virgil); amplerus non aspernata minores, |. 336: cf. thalamos minores, St. Ach., 1. 90; latrantibus undis, |. 361 = Sil. Ital., iii, 471 (which Joseph could not have read); dux aliis, milesque sibi, |. 395: cf. miles in hoc, Libyco dux primus in orbe, Luc., vii, 223; thoracisque moras, |. 429: cf. loricaeque moras, Aen., x, 485; fit fuga, i, 432, vi, 344, 652 = Ov., Fast., ii, 497; lucro tactura fuit, |. 484: cf. par iactura lucro, St., Theb., x, 513 and ex aliena i. lucrum, Vulg., Acts, xxvii, 21 (the idea is quoted five times in the lexx. from the Digest). Book IT
Pyliam senectam, 1. 73 = Mart., viii, 2, 7; redimit moras, ll. $0 f. = St., Theb., vii, 1389; si iusseris, ibunt., 1. 133: ef. si i., ibit, Juv., iii, 78; sibi com- modus uni: cf. mihi c.u. Hor., Ep., i, 9, 9; 1. 186 is a quotation of St., Theb., ii, 429; speciale bonum, |. 195; cf. Prosper Aquit. in Dresemius’ note; gratissimus error, 1. 222 = Hor., Ep., iii, 2, 140, and Aen., x, 392; nostrum est quod regnat, |. 387: n. e. g. uiuis, Pers., v, 131; rara securis, 1.416 = Hor., Sat., i, 7, 27 (i.e. reward of honesty: Joseph is thinking of Acron’s note); Phyllidas Hysiphilas, 1. 488 = Pers., i, 34; fatis nulla fides, i. 575: fronti n.f., Juv., ii, 8; flos Asiae, |. 589 = Juv., v, 56.
Book III
totaque thuricremis Panchaia spirat in aris, |. 10: cf. Georg., ii, 139; Hesione contentus eat, 1. 40 = 1. 195; mater secunda, |. 206 = matertera: cf. Paulus, p. 121, Lindsay, matertera quasi mater altera; falcatur in arcus, 1. 213: falcatus in a., Ov., Met., ii, 229 and Her. ii, 131; praeceps it fama per urbes, |. 218; cf. Aen., iv, 173; non dabit illa manus, |. 250 = Claud., in Eutrop., i, 367; Cicero . . . non opus est ubi fantur opes, 1. 251 f.: ef. proverb ‘nummus ubi loquitur Tullius ipse tacet,’ or Brunellus: munera cum clamant
e ig i- 1- as es ly | et he 2S, nd n- ps cs est ua vas nd ith ber of ind ruat
68 The Bellum Troianum of Joseph of Exeter
caetera quaeque tacent (p. 101, 1. 15 up, ed. Wright), or as the Planctus Nat. has it (p. 488), wbi nummus loquitur Tulliani eloquii tuba raucescit, also Carm. Burana, xix, 9 f., ubi nummus loquitur et lex omnis tacet; tendentemque manus et laeta fronte uocantem, |. 283: pandentemque sinus et tota ueste uocantem, Aen., viii, 712; aerium suspendit iter, 1. 347: a. sibi portat iter, St., Theb., x, 842; decoctius audet, 1. 433: d. audis, Pers., i, 125.
Book IV
plenis fibris, 1. 205: tumidae fibrae, Claud., 1v Cons. Hon., 247 (a similar passage); |. 219: cf. St., Ach., ii, 196; Balearicus augur, |. 231, from false reading of St., Ach., ii, 419 ‘tortae Balearicus actor habenae’ (cf. Balearis tortor habenae, Luc., iii, 710 — see on v, 311); nimborum pincerna Notus, 1. 295: cf. Anticlaud., p. 338; 1. 12, ed. Wright, Auster pluuiae p.; Vulg., Psalm, exxxiv, 7 (rayiav vorov, Job, ix, 9, is not represented in the Vulg.); agnoscens ueteris uestigia cursus, 1. 331: agnosco u. u. flammae, Aen., iv, 23; Et responsuris inuitant uocibus Echo, |. 517: et r. ferit aera u. Amen, Paul. Pell. (see Dresemius).
Book V
numine laesa, |. 61: n. laeso, Aen., i, 8 (did Joseph read laesa ?); et pietas et clara fides, 1. 64: heu p. heu prisca f., Aen., vi., 819; bibit unda cruorem, 1. 179: b. acta cruorem, Aen., ii, 804; reliquisque superbior umbris, 1. 191: aliisque iactantior umbris, St., Theb., ix, 559; deos in uota fatigat, 1. 199: superosque in u. f., St., Theb., ii, 244; formidabile frendens, 1. 206: f. ridens, St., Theb., viii, 582; contorquet plenum galea caput, |. 253: galeas rotat plenas, St., Theb., viii, 700; tortoris Iberi, 1. 311: Balearis tortor habenae, Luc., iii,710; Touts igne citatius, 1. 328 = St., Theb., i, 92; si septem forte fuissent, 1. 417= Juv., vi, 642; rorantes sanguine ramos, |. 432 = St., Theb., ix, 597; omnia certa timet, |. 443: o. tuta timens, Aen., iv, 298; quam tenui nutant mortalia filo, 511: omnia sunt hominum t. pendentia filo, Ov., Pont., iv, 3, 35; Fortunae munera .. . insidias, |. 512 = Sen., Ep., viii, 3.
Book VI
nil refert quis, ubi, quando, sed qualiter et quid factum, 1. 33: (on facti ratio) refert quis, cui, quando, quae, ubi, Sen., Ben., ii, 16; medius descrescit ager, 1. 43: medium decrescere campum, St., Theb., viii, 397 (in spatium iam crescit ager, 1. 638 below); litis litui testes, |. 167: derivation from Paulus, p- 103, Lindsay; portae maritat (‘fastens’), 1. 367: ferrumque maritai aura tenax, Claud., Id., vi, 38; temone iugales, 1. 320 = Claud. Cons. Prob. et Ol., 1. 4; unica Phoenix, |. 382 = Lact. Phoeniz, 1. 31 (Carm. Bur. 33. 4, 96.3, H. Sept., 155, Delic. Cleri, 747); mille licet mortes hinc inde fatigant, |. 448: mille modis leti miseros mors una f., St., Theb., ix, 280; 1. 449 ‘no room for
t] sl as m in be al $1 th m th or th the Mi the by If hin Ta On are
The Bellum Troianum of Joseph of Exeter 69
more wounds’ spoils an excellent passage, yet is the one line for which Joseph had the fullest classic authority. The idea occurs Ov., Met., iii, 237, vi, 388, xv, 529; Pont., ii, 7, 41, iv, 16, 52, [bis, 1. 344 (where these parallels show Ellis is wrong and the Scholiast right), Luc., ii, 177, ix, 814 (cf. ix, 769), St., Theb., v, 598, Sil. Ital., x, 513, Claud., Ruf., ii, 415, Chaucer, Troil., i, 713 f. (where a gloss refers to Ovid), Eurip., Her. Fur., 1. 1250 (many of these examples from Dresemius); Te, Delos . . . te Bacchica Naxos, raderet, |. 484: raditur Lemnos, St., Ach., ii, 3, and B. Naxos, ibid., ii, 4; breuior manus, etc., 1. 610: cf. St., Ach., ii, 205. On 1. 760, nox fera, etc., see SPECULUM, 111 (1928), 351 n.
IV. Tue Text
The text of Joseph is as pure as that of most contemporary poets, though his extreme boldness makes emendation uncertain: for in- stance, nobody would suspect that the vulgate of i, 125: phariscat, concealed the bold coinage phariseat (‘separates,’ from pharisaeus), as we know from the Paris MS. that it does.' We can judge of the merits of this MS. (saec. xiii, referred to as P.— it leaves off early in Bk v) from Jusserand’s reprint (op. cit.) of Bk i. It is certainly better than the unknown MSS used by the older editors, but is already full of corruptions; for instance it is clearly wrong in ll. 270, $13, 446, 477, 481 (apart from mere slips). Its value is enhanced by the commentary it contains (referred to as =). This, like other com- mentaries in MSS of mediaeval poetry, is rather a puzzle. We know that mediaeval authors who prided themselves on their obscurity or cleverness, glossed their own MSS — e.g. the vain Liutprand and the enigmatic Abbo (author of the Bellum Parisiacum). Possibly the custom was fairly common. In the case of Chaucer, certain MSS of Troilus (see Root on v, 799, and art, cit.) quote in the margin the passages of Joseph imitated — with the very corruptions read by Chaucer in his copy (e.g. Calydonius heres for heros, Jos., iv, 349). I feel confident that the information came originally from Chaucer himself. (The same applies to some glosses on the Nun’s Priest’s Tale — see K. Sisam’s notes to his edition, p. 59 ad fin, Oxford, 1927.) On the other hand neither the Chaucer glosses nor those on Joseph are always right: I conclude that the nucleus proceeded from the
1 Pharisaeus (adj.), ‘separate’, occurs in Carm. Bur., 173a, 2.
| ‘ ; | t n r
70 The Bellum Troianum of Joseph of Exeter
author, and was expanded by commentators with varying degrees of knowledge. An instructive case is Troilus, i, 713, which is glossed quaere in Ouidio — the reference to Ovid would be Chaucer’s, but the scribe could not find it: see on vi, 449, pp. 68, 69 above. Certainly the information in > is often of such a nature that it could only have emanated from Joseph himself — e.g. on phariseat (p. 69 above). Sometimes it is right when the text is wrong, e.g. Il. 478, 495; at other times it is itself wrong, e.g. on Il. 39 (a mere slip), 88, 94 f., but on the whole it is better evidence than the actual text. The other MSS which appear to be known are one in the Bodleian and one in the Chapter Library at Westminster Abbey. Probably a collation of the three MSS would settle most of the difficulties with which the text abounds. I have, thanks to the skilled help of Miss E. G. Parker of Oxford, ascertained the reading of the Bodleian MS. (Digby 157, here called B) in some 150 places where I suspected the vulgate.'
B supports the Paris MS. in Book i, ll. 11, 38, 44 (sacerdos), 68, 80, 92, 95, 105, 125, 197, 228, 248, 257, 268, 365, 375, 397, 438, 453, 471, 477.
It supports the editions against the Paris MS. in Bk. i, 23a (maronem), 32f-g (same as Delphin ed., p. 640), 167, 174, 201, 266, 313, 446.
In i, 495, it has indulsit with 2.
It supports Dresemius’ conjectures in iii, 36; vi, 246; vi, 554; vi, 558; vi, 885.
It supports my own conjectures in the cases specified below.
It offers the following new readings: ii, 121, dampnatis; 162, producte; 282, in uultum facie (for in florem cultu); 415, superbum; 579, nutricibus. iii, 130, armata decurre manu; 150, iugo (for neci); 293, o for non; 342, dur augure; 360, spectanti; 370, facie. iv, 116 uox teneris . . . locis; 146, ceu russa iacens; 157, aspectum (so Westminster MS.); 158, flauicies (so Root); 192, arcus; 212 f., at dephica missus fata exoratum; 348, sepulti. v, 89 (sup- ports v.1.); 110, augentibus ipsis; 292, et . . . ueniunt; after 530, five lines (Jusserand, op. cit., p. 96). vi, 556, uigilantibus; 557, mento; 819, famulate; 895, ternusium.
1 The MS. is well written on parchment in small folio, and assigned in W. D. Macray’s catalogue of the Digby MSS to the twelfth century. It contains: (1) Bernardus Silvestris de Creatione; (2) Architrenius; (3) Poema de Bello Troiano ex Darete Phrygio. Miss Parker writes, ‘No. 3 is written in very small neat hand, rubric notes in margin occasionally, e.g. Oratio Veneris. I see only one gloss, uel sceptri (to i, 225, quam regni iactura mei), written perhaps in the same hand but darker ink.’ It contains the whole poem. _
The Bellum Troianum of Joseph of Exeter 71
In some 60 other places examined it offered nothing new. It will be seen that B agrees fairly closely with P; in fact in Bk i, where we know the readings of P, it offers no new reading of any impor- tance. It is to be feared that both P and B are by no means free from serious corruption. I conclude with a number of conjectures, on passages where no recorded reading seems satisfactory:
Book I
ll. 42 f., I think the vulgate is right, with litata (P) for litatam; trans- late: ‘into whose charge that kind father and lofty pinnacle of the church [Becket] would wish to come all he gained by his noble martyrdom and the peace he brought by the sacifice of his life’; 1. 73, contenta: contempta?, a constant confusion in MSS; 1. 83, stat dubius nouus hospes aqua (aquae, P): perhaps dubiae . . . aguae, with Joseph’s favorite rhyme; 1. 201, Cerauneis: Therapneis (= Castor and Pollux, as in Statius); 1. 380, dicamne uiros? is a parenthesis: delete query after pudet; 1. 299, aure: ore (consilio, =); 1. 377: punctuate, Danaos: quae iam mora, segnes?; 1. 380, traxere: cinxere (enixere B); 1. 387 read inhospita. (so B); 1. 450, effractis: effractam; cf. 1. 410 (so B); ll. 478 f., read contempta, i.e. contenta, and senum (so B): see =:1. 502, nec... . nec: nunc . . . nunc (so B); |. 513, canicolam; canticolam, ‘song-lover’? (cancic. P, cancric. Root, ‘ferwentem,’ = — implying céniculam? cancicola [sic!] B); 1. 526, alterius: alternis; 1. 535, naualia: muralia?
Book II
1. 10, fuit; fui (so B); 1. 94, syrtis (musical instrument): sistris (so B); 1. 104, docilis: dociles?; 1. 174, wel omni: uolenti., 1. 178, cum leuis ascensu: tum 1. assensu (assensu B); |. 250, ausus: ausit?; 1. 338, haeccine forma: haec mihi f. (no question); 1. 394, tractat: tractet (a dependent question); 1. 459 positura: passura; |. 535, seuerum: fidelem? or better, serenum (hardly seni- lem, = anilem, of the nurse Beroe); in 1. 536, the query should come after quid, |. 588, aspera: pocula?
Book III
1. 51, augustat: angustat? ‘reduces, constrains, to practical utility’ (so B); 1. 52, praemensa futuri seems impossible, in spite of mensa futurum in 75: praesaga?; 1. 139, facilis: faciles; 1. 151, read reddite fatidicos, wictores ensibus, enses (‘ignore baleful prophecies’) ; 1. 156, proficit: profuit; 1. 207, ce- lebrem: celebris? (see on style, above); 1. 208, wocabat: uacabat; 1. 230, actus in armos: auctus (so B), cf. iv, 43, 65; ll. 248 ff., delete comma after stipatriz and put colon after inuidet, ‘grudges him the chance of asking’; |. 293, non: nunc; |. 303 stridere: sordere (?)
d e y t
S e e . t yf f
i 2, , 3; e; 8. ); =] e; de er en
a
The Bellum Troianum of Joseph of Exeter
Book IV
1. 29, certant: certat; |. 66, acie: facie; 1. 106, iudice: uindice, cf. 169 and see examples of style (so B); 1. 129, dandi: fandi; 1. 158, planities is probably sound, ef. Planct. Nat., p. 482, Wright, p. frontis and p. menti, Antiel., ibid., p. 283, p. menti;! 1. 199, ae teneri scintillat: at (so B).. . titillat (‘her anger was mild, but her love passionate’); ll. 212 ff., corrupt: for ad, perhaps ac; for missum, missi? refugas, refugum (‘unwilling’)?; 1. 393, fluit:
furit (fuit B); 1. 395: full stop at end: delete colon in 396; 1. 506, wincet: :
uincent (so B); 1. 514, lugens: ingens. 4
a
Book V al
a
1. 57: read et precibus mens tarda negat; quin, aspera rerum (et = ‘also’: a negat = ‘refuses to obey’?); 1. 76, et: at (or semicolon after erit); 1. 81, semicolon after age: delete comma at end; |. 110, see above, p. 58; 1. 287,
bellis: belli; 1. 292, et: ut; 1. 341, mutat: motat?; |. 372, uiscera lapsa: apparent- ba ly uiscere lapso; |. 422, bellis: stellis (so B); 1. 427, passusne metus: pressoque
metu? ca
Book VI ce
1. 8, at: ac; 1. 11, non: sed (non from previous lines)?; 1. 97, praerepta: te
praecepta (so B); 1. 174, cum: dum; |. 254, hine: huic (B’s reading might be
either); 1. 279, profundo: -a; |. 688, sic: si; 1. 769: punctuate incurrit fatum
fati fuga: sola . . .; 1. 771: delete comma: passis depends on fas(est) and 7
dash after dolores —the next line is in apposition; 1. 836, pars: Paris
(so B); 1. 861 wulgo: uulgi; 1. 895, Cernusium: Chersonésum (cf. Dares, 43): dei
Joseph could not know the quantity. del
det
V. VOCABULARY die
t
The following list contains the words which seem to me most »
striking and characteristic in Joseph; naturally it is only a selection, F:
but I hope it is representative and omits nothing of real importance: tdi
for a real index of Latinity every line would have to be quoted at dos
least once, often twice or three times. It will be seen that it is a euel
fair length for a poem half the length of the Aeneid. The references ~
to classical authors help to show Joseph’s indebtedness, but in a few vi
cases I could only exemplify a word from an author unknown to a
a
Joseph, e.g. Silius. ffau
of
1 mentis in Wright. The original is no doubt Sidon., Ep., i, 2. 2, p. frontis.
|
The Bellum Troianum of Joseph of Exeter 73
A dagger (f) prefixed to a word shows it does not occur in the dictionaries based on Forcellini — at least in anything like the re- quired sense. (This does not include metaphorical uses, which in
Joseph are endless.)
angustia i, 162
apex rerum = Deus ii, 40
Argolidae v, 386
arteria (‘throat’) ii, 96
tarticuli (‘limbs’) iv, 95
arx capitis vi, 584 (Stat.)
astus vi, 309 (Sen. Sil. Prud.)
ausus, -iis i, 290 (ii, 2507), v, 53, 60, 69, 233, 244, vi, 693, etc. (v. rare)
baiula lethi (Minerva with Gor- gon) ii, 253
causa (‘pretext’) vi, 704 (un- class. use)
cella secunda (of brain) vi, 255: cf. thalamus
tchoraules (‘conductor’?) ii, 102
consultus, -iis = consilium (doubtful form and meaning in Class. Lat.)
cura tridentis Neptune, ii, 426; cf. cura harae (Eumaeus) Ov. Her. i, 104.
deitas ii, 542 (Prud.)
deliciae lethi ‘fastidious choice’ i, 991 (cf. Juv., x, 291)
deploratus vi, 402 (perhaps a
verb)
dies (‘daily bread’ ?) iv, 507 (cf. the M.A. use of diaeta, Spec- ULUM, 111 (1928), 367)
dispendia Asiae (‘wide Asia’) iv, 20; d. noctis v, 38 (d. stluae, Luc.)
tdistitium v, 230
dos abs. (‘talent’) ii, 98
euenta (‘the past’) iii, 144
exul as adj. ii, 430; exul bellum (‘war banished from mind’) vi, 863
famulatrix fama i, 248 (Sidon.) tfauillae (‘ames’) vi, 838 ffauus (‘sweetness’) i, 464 (v. often in M.A.) mira (‘a wonder’) v, 99
NOUNS
ffimbria (‘ornament’) (pau- onis) iv, 96 (cf. SPECULUM, II (1928), 368)
(‘offspring’) vi, 841
(Ov. jfrons (‘leaf of book’ ?) ii, 401
genae (‘eyes’) vi, 820, 834: cf. diesco. (Ov.): (face) ii, 236; iv, 187
tin gyrum ii, 456
hiatus mortales (greed) i, 56 hyems, met. = misfortune, v, 60 (v. often in M. A.)
fiuga crurum v, 264
flanugo (of laurel) vi, 408
Latonigenae iii, 258 (Ov.)
leno oculus iii, 244 (i.q. adulter) iii, 309 (cf. SpecuLUM, 111, (1928), 369)
libra (‘judgment’) ii, 241
libramen (met.) i, 341, iv, 190 (Claud. Mamert.)
fliuor (‘pride’) iv, 137
maculae of helmet (?) v, 165
manus suprema (‘final task’) v, 477. m. campestris (‘battle on plain’) vi, 565
fnecesse as noun iii, 14, vi, 257 nutrix opera (‘providing daily bread’) iii, 351
obsequium is (‘pliancy’) vi, 907
pecten ferratum (‘portcullis’) i, 482
peplus fraxineus (‘banner on ash-pole’) vi, 583
pietas (Dares’ misericors) iv, 162, iv, 217: cf. pius. (Suet.)
pincerna Notorum iv, 295, v. imitations. (late prose)
fplausus (‘dancing’) i, 504 (2) ii, 79 (93?)
praeceptrix tuba vi, 166
praedatrix i, 94 (these words in -tor, -trix are characteristic of Prudentius)
proreta i, 191 (Plaut. Gloss.)
pudor (‘thing to be ashamed of’?) ii, 192
tpyralia (‘chimneys’) (camini i, 499
tpyropus (ornament of armor) v, 1
repetitor iii, 115 (Ov.)
respiramen (‘respite’) i, 530
(‘windpipe’ in Ov.)
rupes: non (laesit) rupibus au- rum (‘did not insult the gold with precious stones’?) i, 82
tscansilia (‘stirrups’) vi, 262 (Dresemius has an interesting note on the neo-Latin words for stirrup)
jsigillum (‘epitaph’) iv, 488
sititor belli iii, 65 (Apul.)
stipatrix turba iii, 249
fstrata (‘pavement’) neut. pl. vi, 176
tstylus (‘spear’) vi, 342
sylvae trabalis robur (a spear) v, 208
teliger = Cupid ii, 421 (Sen., Trag.)
tempestas uulgi (= popularis aura) iv, 29
ttenor Libycus (Syrtes) iii, 294
thalami uitales iv, 194 (Specu- LuM, 11 (1928), 371, and Planct. Nat., ed. Wright, 1, 453)
tractus (‘length’) iv, 187
tuba Nestoris (voice) i, 136
ultrix (‘claimant’) ii, 429 (anal- ogy of uindico, uendico?)
umbo Maleae iii, 412 (= St., Ach., i, 408), v, 313
fumbramen i, 429
tuespera (of morning twilight)
v, 99 uestigia (‘feet’) vi, 619; cf. Aen., v, 566
j y ut l, | — 7,
1: is ): n, | at
74 The Bellum Troianum of Joseph of Exeter
adgemo with dat. vi, 374 (Stat.)
admissi (equi) vi, 639
aggerare iram v, 498 (Stat.)
alternant caedes v, 355
augusto iii, 51 (Stat.)
aspero iv, 396; vi, 489; a. cach- innum vy, 144 (not in class. verse)
augesco iv, 165
candesco iv, 183, 454 (Ov.— rare)
canesco mente i, 19. See Sprc- ULUM, 111 (1928), 366; add Liutprand, Ant. i, 44, Greg. Nys., Enc. Bas., 1, modtbs; Architr. ix, 200 (p.
fcausor (‘blame’) i, 211
clareo iii, 366
concido with dat. ‘fall with’ v, 473
coniecto (‘guess’?) v, 397
corono (‘surround’) vi, 436: see on Style. (cf. suspendo)
crudesco (‘be [grow] cruel’) ii, 264, vi, 607, 643 (Virg.)
decaluo i, 404 (Vulg.)
dedisco i, 278
tdefoedero ii, 436
deferueo i, 213 (Vulg.)
tdegladiatus vi, 649
delambo ii, 256
tdelinita (deflorata >) i, 150. linio = ‘smear,’ ‘defile,’ in Med. Lat. So Brun., p. 129; cf. Joh. Monach., ed. Huber, p. 48. Salomon et Marcolfus, ed. Benary (Gloss.). Linio, -onis = sordes, Ad. Vien., 104
tdemensus (‘ disproportioned P ?) iv, 144
deplango iv, 493 (Ov.)
tdepotuisse, opposite of potu- isse i, 281
oe i, 291 (Stat. — only in part
despumo i iv, 288 (Luc.)
detersus (‘clear’ of sky): iv, 311, v, 384, 6, 854
tdetitulo (‘deprive of glory’) ii, 349
Tdiescit toto ore ii, 608; cf. gena- rum dies =‘eye’ 369: (Ant. 1. 160, and pp. 342,
345, ed. Wri ht: ad fin. — wi
noctesco)
VERBS
diffibulo (met.) v, 131. lit. in Stat., met. often in Anticl.; Architr. ix, ch. 11; ef. Specu- LuM, 111 (1928), 373
dignor (‘honour’) abs. i, 529
discrimino (‘separate’) (lit.) vi, 192
effemino (met.) vi,
emasculo (met.) ii, “at (lit. in Apul.)
emereor = mereor, i, 518, ii, 539; with inf. ii, 483 (as Ov., A.A., iii, 410)
temungi (minut 2) i, 244 (cf. Alda, 415)
erigo ictum (‘rise to strike’) vi, 643
exarmatus vi, 324 (Luc. etc.)
exeo with acc. vi, 782 (Virg.)
textenebro i, 303 (had Joseph somewhere misread eztere- bro?)
texurens nuntius aures vi, 539
facit ad ‘supports’ v, 16
ffalcatur in arcus iii, 213 (falca- tus Ov.). Joseph often coins verbs from adjs. in -atus; see below.
fallenda senectus ‘disguised,’ opposite of fingenda, ii, 546
tfastidio = fastidio afficio i, 89, 198, 225; cf. fastiditior.
fatigat cibos (‘won't eat ’)ii, 154
faxo with subj. vi, 295
febricitent vultus vi, 82
flauesco iv, 164
furor (‘conceal’ body by dress) v, 33 (Sen.); furatur uox blaesa sonos iv, 52
generat uoces (‘gives deriva- tion’) ii, 400 (= euoluit nom- ina, ibid. )
gemisco vi, 678 (Claud.)
germinat in sensum (‘comes into thoughts’) vi, 770
hilaro i, 250 (Cie. Ov.)
ignescere (of eyes) iv, 397
impingo (equos) v, 244, 333 (Statian word)
timpraegnare vela i, 107
tincendit decor aliquem (‘ glori- fies’) iv, 79
infamo vi, 269
inserpit (death) labris iv, 440 (Sta
insignio iii, Ving.)
instrepo iii, 439 (Virg.)
integro i, 476, v, 52 (Virg.; oft. in Stat.)
interfatur ii, 186 (Virg.)
interiaceo vi, 680 (Claud.)
interuenio vi, 232
inuergo iii, 445 (Virg.)
laborat in cumulum apex i, 537
laxatur in pectus (‘broadens out in the chest’) iv, 110
libro i, 298, 385 (v.1.); ii, 1%, 241, 581, 584; iii, 135, 225, 230; iv, 49, 74, 94, 129, 156, 196, 315; v, 2, 31, 172, 355, 406, 501; vi, 907
liuesco ii, 463, iv, 152 (St., Th, iv, 58)
magistro (‘teach’) iv, 332 (mo- gistero in Festus)
tmedians with dat. (‘interver ing’) i, 505
medicatus murus i, 477
mergo (‘spoil,’ ‘ruin’) iv, 14, 201, 206
fnegauit furta (‘refused [to wit by] guile’) v, 11; vi, 716. negat uicisse (‘refuses’) v, 22
jneutrata genas mixto nitor iii, 361 is explained by Pod. Noua, 572, neutro colore
obliquo dolos iv, 123; cachinne vi, 963; (‘insinuate’) v, 289; (‘look askance at’) with ac iv, 138. v, 219 (Statian word
obuio iii, 437 (Jerome)
opimo iv, 117 (Auson.)
tpennifico v, 203 (see on Stviz
fperegrino, met. of anythin strange (for deponent) i, 7, 239 (526); cf. Henr. Sept iii, 70, Anticlaud. i, 25
equo = deturbal,
tphalero i iii, 394, iv, 102
tphariseo i, 125 2)
plano vi, 384 (Corip )
tplebesco vi, 348 (Anticl, i, 118 165)
praecipio (‘anticipate’) vi, 655
praeradio iii, 170
tr pr pr qu rig rol ror ‘ tsa adt 384 tac e alip t anh ard fe astr auc! tau cast ‘ in vere cita us com iii A th coni cons 7 cont corn 64 cring teun iv, daps vi, 299 pe discr iv, disci: dissu is j
The Bellum Troianum of Joseph of Exeter
raeresero iv, 69 ee (‘hide’) vi, 218 (late
rose prsenelo iii, 162 (Claud.)
quadratur in artus (Dares’ quadratus) i iv, 125
58 (Cie. ) rorantia ry ii, 495: rorant
clypei vi, 2 tsancio (‘reconcile’) ii, 540
funera (‘of adulterer’)
(‘big’) i, 473; excessus 3, 238
alipedes (of horses) v, 338 (8 times in Stat.)
anhela litora; (?); iv, 521
ardua pingi gloria iv, 79; a. in fastum iv,144
astrifer i, 580 (Stat.)
fastriuagus vi, 115
auctius (adv.) iv, 85 (Hor.)
fauricomum uellus i, 185
— (adv.) iv, 198 (adj. in A
eerea oscula (of the dead) vi 199 —_ (adv.) v, 328 (adj. in
complutus (met.) i, 471, ii, 76, iii, 246 (Jerome in lit. sense: Anticl. praef. 8, and often, in the metaphorical)
coniugus ii, 339, iii, 278 (Apul.)
consonus (‘consenting to’) vi, 71
contractior v, 104 (Cic. etc.)
cornipedes (of horses) vi, 176, 640 (often in Claud.)
crinalis aurum iv, 178 (Ov.)
teunalis Mars (‘fight in cradle’) i, 263; honos (‘noble cradle’) iv, 463
dapsilis i, 257 (rare in class. period)
denoctior i iii, 433 (Pers.i, 125) discretus = diuersus i, 251, iv, 32 discinctus (‘affable’) iv, 116 dissutus (met.) vi, 816 (the word is in Jerome)
sedet (‘it is settled’), with inf. iv,
socio, intr. v, 180
soporo iv, 520
spargo (of Fame) i, 51, 111, 300
spatior (of sailing) iv, 324
stagno v, 374 (Ov.)
succingo (‘make small’) iv, 187 (see the passage quoted p. 56)
summitto (‘press down’ ?) ii, 219
superbio with inf. (‘scorn to’) vi, 268, 489 etc.
suspendit aquas (of river) i, 165; cf. i, 156, 523: s.iter i, 521,
ADJECTIVES AND ADVERBS
felinguis nidus (‘silent’) 431; cf. iv. 41
ffalsigraphus ii, 485 (in Aegid. Corb., i, 80, of the corruption of a MS.)
fastiditior (“more fastidious’) i, 77
fossor ictus vi, 561
fractus (‘aged’) iii, 19
fusus (of limbs) ‘mighty’ iv, 99, 116
(‘abundant’ ?) iv, 104 tglandifluus (of sling) vi, 225
therilis = superbus iv, 97, 156 hospita, neut. pl. (‘home’) vi, 573
humilis manus (‘feeble’) v, 497
impatiens with inf.,= non pa- tiens v, 446 (Claud.)
Inachus (Greek) v, 388 (Stat.)
inausus (‘undared’) iv, 135 (Virg.)
findecoctus iii, 181 (v. decoc- ti
‘ior
finexpletus (‘incomplete’) iv, 180, v, 149
inexplicitus (‘concealed’) i, 210. (Stat.)
infraenus v, 333 (once in Virg.)
inoffensus iii, 89
intensior v, 41 (Sen.); -ius v, 98 (only Fronto)
interflua labes (‘intervening’) iv, 191 (the word is in Pliny)
iaculabilis hasta vi, 227, 501 (Ov.) ;
75
iii, 347 (cf. 356), iv, 189; 8. ceruice genas (‘face’) iv, 137; medium suspendunt (= habent) Hectora v, 407;
vi, 176
tepeo (‘am hot’) iv, 393; (‘am cool’) iv, 204 (see SpecuLuM, 111 (1928), 376)
tinnio (of bird) vi, 306 (Calp.)
ululatum ‘uale’ iv, 494
unio iv, 270 (Sen., Trag.)
turo aera clangoribus i 334
uerno (artubus) iv, 113; cf. 182, 453, etc. (Ov.)
iunctissimus (‘ very close’) v, 430 (not used in local sense)
Iunonior i, 268 (v. Specutum, 111 (1928), 359; add Carm. Bur., ii, 5, 8)
latrantes undae (Scylla) i, 361 (Silius)
liquidus i, 286
longum = diu i, 313, iii, 365, vi, 540: longa (n.pl.) vi, 631
fluctificus vi, 370
lunatum robur (horns) i, 292
Iuppiter = Neptune
minax with prolative inf. iv, 136
mollius meminit (‘worse mem- ory’) ii, 402
tmonstriparus i, 414
mortalis rogus vi, 465
myrtea coma iv, 1il
nudos hiatus (‘toothless’) v, 435 numerosus Proteus (= varius) ii, 444
obtrusus (‘obstructed’) iv, 450 (Prud.)
focciduum lumen (‘downcast eyes’) iv, 56; coma vi, 228
fomnificus (zavodpyos) i, 52, ii, 435
tpius (‘kind’) ii, 418, 461 pretas)
-— presagia (‘fulfilled’) iii, 15
pomposus ii, 487 (Sidon.) populoso pignore (“many dren’) ii, 1
tproclivius (‘lower’) i, 511
j = f 4 i, 587 ns out 1%, ), 156, 355, Th, 2 &§ ervel- 148, negat 2 &g nitor y Pod. e | hinnes v, 289; | ith act. word nything 4, 73, Sept vi, 658
76
productior vi, 777 (Colum., Prosp. Aquit.)
progressior iii, 168; progressus = procerus iv, 142 (Tert.)
roiectior aluus iv, 139
rona abs., (of mind ‘grovel- ling’) i, 324
tpugil v, 321, vi, 492, 688
putris nubes (‘cloud of dust’) vi, 217
tpygmaeus (‘small’) i, 538
remeabilis (‘returning’) vi, 536 (Stat.)
retro datus (‘falling back’) v, 491
riguus v, 377 (Virg.)
frorificus vi, 853
tseductius = remotius ii, 212
separ (abl. separe) vi, 466 (as Stat., Th., iv, 481)
sequestra pax iv, 351
specialis iv, 91
spirans uulnus = spirantis v, 474
Wycaceston Boys’ Scuoot,
LeIcEsTER, ENGLAND
The Bellum Troianum of Joseph of Exeter
succinctum caput (‘small’) jy, 51; sura iv, 77
surda manus (‘not obeying’) vi, 510 (cf. Stat., Th., ix, 864)
teliger (Cupid) ii, 421
thuricremus iii, 10
tricuspis vi, 264 (Ov.)
trisulcus (igne trisulco) iv, 382 (Festus, p. 480, Lindsay)
uenialis v, 60 (Macr. Sidon.) fuentripotens ii, 80
MEDIAEVAL ACADEMY EXCAVATIONS AT CLUNY
By KENNETH JOHN CONANT Research Associate in Archaeology of the Mediaeval Academy of America
Vv
Tue Date or tHe AMBULATORY CAPITALS
HE technical studies carried on in connection with the Acap-
EMyY’s excavations at Cluiiy in the summer of 1929 brought forth one conclusion of very great interest, which permits a final pro- nouncement on one of the most hotly debated questions in the history of mediaeval sculpture. The eight beautiful limestone capitals from the apse of the abbey church, now preserved in the local Museum at Cluny, were taken down from their shafts in order to be copied in plaster and photographed. While they were on the floor and thus ac- cessible to close study, the writer observed that the sculpture of the sides of five of the capitals is carried over the upper edge and onto the top of the block for a short distance. This peculiarity would be of little importance but for the fact that such capitals in Bur- gundian Romanesque architecture invariably carried an overhang- ing moulded impost block, the dimensions of which were fixed by the outer angles of the capital. It follows that none of the upper surface of the capital was accessible to the sculptor’s chisel after the impost block was set, and therefore that the great capitals were carved be- fore the block was put into place. Messrs Porter, Oursel, and other archaeologists, including the present writer, had long believed this to be the case, without the categorical proof which is now in hand. The writer observed other capitals still in situ on the building which could not have been carved after they had been put into place: two on the minor portal which once connected the nave with the cloister, several in the Chapel of St Gabriel in the square tower of the tran- sept, and a number in the transept itself, including one at the east on the exterior which strongly resembles one of the ambulatory capitals (SpecuLuUM, Iv (1928), 25; Bulletin Monumental, txxxvu (1928),
77
)v ‘ing’ , 864) i r, 382 | y) n.)
|_|
78 Mediaeval Academy Excavations at Cluny, V
55 ff.). The certainty that the great capitals were carved at the time of setting completes the proof that this procedure was the one followed at Cluny, and it puts a great burden of proof on all who do not believe that the capitals were carved before 1095, when the sanctuary of the church was in condition to be dedicated by Pope Urban IT.
All archaeologists must, in the light of the new observation, ad- mit that the carvings date from shortly after 1088, when the church was begun, or shortly after a fall of vaulting in 1125, which, accord- ing to some writers, caused a far-reaching restoration, but before 1130 or 1131, the date of the definitive consecration of the church by Innocent II. Only these two dates are possible.
For the stone blocks from which the capitals were carved the proof is already complete, because all of the upper and lower sur- faces are worked with vigorous strokes of the hatchet in the eleventh- century manner found in the oldest parts of the building, and already modified when the facade was built early in the twelfth century, as the excavations of 1929 have shown. If, therefore, the carving was done after 1125, it was upon blocks prepared a generation earlier, which is in itself unlikely.
We now know much more about the construction of the church than formerly, owing to the intensive study in connection with the excavations, and it is therefore possible to comment with much pre- cision on the series of texts which give us our information about the building.! The texts come at short intervals and give a satisfactory account of the progress of the building. They prove that the dis- aster of 1125 cannot be magnified into an occasion for a general re- construction, and therefore that the date 1125 for the capitals is inadmissible.
I. 1049. Tuer or St Hucu
Anno ab Incarnatione Domini 1049. Sanctissimus et clementis{simus] Pater Odilo, Siluiniaci degens, uigila Circuncisionis Domini, mortalis uitae adeptus est finem. Quem locum praecessoris sui S. Mayoli insignem cor-
1 The punctuation and capitalization are modernized; the modern use of ¢ and u for i, u, j, v is adopted here.
| I ( t d cl a di eX fo m
Mediaeval Academy Excavations at Cluny, V 79
pore, ipse etiam nobilitat proprii corporis tumulatione. . . . Pater Hugo electus. — M. Marrier, Bibliotheca Cluniacensis (Paris, 1614), col. 1621;
from the Chronologia.
St Hugh of Semur was twenty-five at this time; nearly sixty-five when the great church was undertaken (late in the year 1088), and eighty-five when he died.
II. Asout 1080. Lrerrer or Atronso VI to St HucH
De cetero uero sciatis me ad presens in pacificandis Yspanie urbibus ualde adtentum esse, que postquam, quod, Deo annuente, in proximo fu- turum est, mihi adcline fuerint, ecclesie quam edificatis auxilium faciam et uoluntati uestre summopere adquiescam. Nunc uero per domnum Sigui- num, uirum omni ueneracione dignissimum, decem milia talentorum uobis transmitto. De Nazara uero, illud quod domnus archiepiscopus precipiet et auctoritas domni pape confirmabit stabilitum erit. Orate pro me, pater et sanctissimum gregem uestrum, ut pro me oret, commonere facite. Domno uere Siguino de his que uobis de mea parte dicet, tamquam mihi si presens adessem, credite. —- A. Bernard and A. Bruel, Recueil de Chartes
(Paris, 1888) 1v, 697.
This letter is important because it seems to give the financial basis of the church, and to show that its construction was under con- sideration for some time before 1088. It is fair to state that the Chapel of Notre-Dame de |’Infirmerie, dedicated at some time be- tween 1078 and 1088, may be intended; or both may have been un- derstood. The vast enterprise of studying the plans of the great church, the preparation of the site, the acquisition of quarries, the arrangements for transport (the quarries were two or three miles distant and the ambulatory columns were brought from Rome), for excavation (the foundations were carried to gravel, at fourteen feet), for the various shops, for the shelter of the crew, for the timber and machinery needed, could easily absorb several years.
III. 1088. Tue ‘Founpation’ or THE CHURCH Anno 1088. Fundatio huius Basilicae 2. Calend. Octobris. — Biblio- theca Cluniacensis, col. 1621; from the Chronologia. This may apply to the beginning of the construction, or to a cor- nerstone ceremony of which we have no knowledge.
} e h y e r- ly AS as r, *h he e- he ry is- is us] Lae or- Jy
Mediaeval Academy Excavations at Cluny, V
IV. 1090. RervENUE FoR THE WorK
Quia uero placuit omnipotentiae creatoris in regno Hesperidum roborare solium meum, ego Dei gratia Rex Adefonsus, sicut heres paternae digni- tatis, ita quoque bonae successor uoluntatis, pactum fraternae societatis cum meis Cluniacensibus inii, statui, firmaui, censum quoque largitatis paternae duplicaui, duo milia mancales in censu annuali Cluniaco per- soluens. —S. Baluze, Miscellaneorum Libri Septem (Paris, 1700), vi, 472; from the Chartularium S. Hugonis.
Here King Alfonso VI of Spain doubles the previous annual con- tribution to Cluny, probably because of the work on the church. The golden mancalis is mentioned in Du Cange, but no value for it is given.
V. 1095, VIII Katenps or NovemBer. THE First OF THE CHURCH
Praeterea rogatus a Domno Hugone ipsius monasterii uenerabili Abbate altare maius nouae basilicae, astantibus plurimis Episcopis, monachis, clericis quoque, ac plebe innumerabili, in honore resurrectionis Domini nostri Iesu Christi & beatae semper uirginis Mariae sanctorumque Aposto- lorum Petri & Pauli ac protomartyris Stephani deuotissime consecrauit vit. Kal. Novembris, indictione m1. [sic] & praecepit ut in ipso die eadem basilica oportuno tempore dedicaretur. — S. Baluze, op. cit., p. 475; from the Chartularium.
The pope asked that the final dedication take place on the same date as the consecration which he performed. This, as we shall see, actually occurred.
VI. 1095, VIII KaLenps or Novemser. Tue First Depication OF THE CHURCH
Anno incarnationis Dominicae millesimo nonagesimo quinto, indictione m1. 8. Calend. Nouemb. Domnus & uenerabilis Urbanus Papa secundus sacrauit altare primum, & maius, noui nostri Monasterii in honorem Dei, in memoriam beatorum Apostolorum Petri & Pauli. Sacrauit etiam per se & altare secundum Missae matutinalis. Lugdunensis autem Archiepis- copus Hugo, Pisanus Archiepiscopus Dabertus; Episcopus Signanus Bruno eodem die in ipso Monasterio iubente Papa tria in tribus primis can- cellis sacrarunt altaria. Tunc Papa inter sacrando, Missasque agendo,
80
P
is
b
th
th
ar
se
Mediaeval Academy Excavations at Cluny, V 81
post alia salutis hortamenta coram Episcopis & Cardinalibus multorumque personis, huiuscemodi sermonem habuit ad populum. Sancti Patres & Maiores nostri Romani Pontifices, qui sanctae Sedi Apostolicae praese- derunt, ex quo locus Cluniacus institutus est ab initio, & Monasterium istud fundatum, tam locum hunc quam Rectores, uel habitatores eius pro-
nsius dilexerunt, fouerunt, & curauerunt attentius. Et merito. Nam pius ille Willelmus, istius olim Monasterii institutor, nulli alii Aduocato, nulli Patrono, nulli Regi, uel Principi curam ipsius, tutelamque commen- dauit, nisi Deo & beato Petro, eiusque Vicariis Roman{is] scilicet Ponti- ficibus. Quorum numero uel ordini diuina me dignatio, licet indignum as- sociauit, me olim Monachum Prioremque Monasterii huius sub domno ac uenerabili Hugone Dei misericordia adhuc superstite & beneualente. Igitur sicut Pontifices summi ante me succedebant sibi in Apostolica sede, successerunt etiam ad tuendum curandumque propensius locum istum. Verumtamen nullus eorum per suam corporalem praesentiam locum istum hactenus uisitauit. Mihi uero, sicut in praesentiarum cernitis, id diuina concessit clementia. Denique inter alias causas quae nos ad uisitandas Gal- lias impulerunt, haec prima & praecipua fuit, ut locum istum & congrega- tionem hance speciali nobis cognatione germanam nostra praesentia laeti- ficaremus, nostro accessu ' alloquio iuuaremus, & ad omnem utilitatem uel commodum nostram eis operam impenderemus. Itaque hic uobiscum hodie praesentes, altare primum & maius cum caeteris quae parata sunt, noui huius Monasterii, sacramus, & ad eam quae de eodem Monasterio restat, structuram, uestros animos incitamus. Placet etiam nobis, uobisque placeat suademus (nam hoc ipsum & bonum uidetur & iustum,) huic loco, qui uobis et caeteris Christianis in ueneratione & cura bona habendus est, quosdam certos limites immunitatis ac securitatis, cireumcirca undique assignare, ipsosque limites sacri banni. — Bibliotheca Cluniacensis, coll. 518, 519.
It will be observed that the work was incomplete, but the sermon was evidently pronounced in the church with a great many persons present. The idea that the church was hardly above ground in 1095 is therefore erroneous. In modern times the altar of a future great church is sometimes dedicated, enclosed in a temporary structure; but here the pope consecrated two altars, while the archbishops and the bishop consecrated the altars in tribus primis cancellis, probably the central absidiole and the two on the northern side. The chances are that only such sanctuaries as were completely finished were con- secrated. It is thus more than likely that the ambulatory was al-
re tis tis | m- ch. it ate his, | nini | sto- lem rom ume see ’ jone ndus Dei, | per epis- runo can- ndo,
82 Mediaeval Academy Excavations at Cluny, V
ready complete, and consequently that the carven capitals were in their places. We know that the first chapel on the southern side was furnished before the middle of December, 1095, but it and the remaining chapel may have been blocked up with scaffolding or still unfinished in October. This being the case, it cannot be asserted that sanctuary vault was in place at the time of the ceremony, though it may have been.
The minor transept was probably at least partly finished in 1095. It was set out very accurately at a right angle to the axis of the sanc- tuary, whereas the axis of the great transept is laid out inaccurately. There are signs of an interruption in the bay just to the west of the still existing west wall of the south minor transept. Here there is a series of heavy cut stone blocks arranged like steps, running through the aisle wall. All indications point to the unity of the work to the east of this point, but it would be too much to affirm that the minor transepts were vaulted in 1095. When their altars were dedicated is not certain, but there is no reason to assume that they waited until the ceremony of 1130 or 1131 — or even until 1115, the termi- nus ante quem for an altar to the west of the great transept.
VII. 1095. InscripTioN IN THE DESTROYED CHAPEL oF St JAMES (AT THE SOUTH END OF THE AMBULATORY)
Hoc altare constructum est a Domino Dalmachuio Sancti Iacobi Apostoli Episcopo, & istius loci Monacho, in honorem eiusdem beati Iacobi, ac omnium Sanctorum in quo continentur plures reliquiae Sanctorum, quorum nomina non sunt nisi de duobus uidelicet Emiterii & Zeledonii Martyrum. — A. de Yepes, Coronica General de la Orden de S. Benito, vi (Valladolid, 1617), fol. 436.
The Coronica was written in the early seventeenth century, and offers no guarantee that the inscription dates from the time of the dedication, but the dedication of the Chapel of St Gabriel is attested to by a similar inscription which is contemporary. Dalmatius had been a Cluniac visitor in Spain. He had a hand in the building of the Cathedral of Santiago (1077-1122 and later). He died on De- cember 13, 1095.
t q | is d te Pp Se
|
Mediaeval Academy Excavations at Cluny, V 83
VIII. Axsovut 1100. DamaGep Inscription IN THE CHAPEL OF
St GABRIEL, IN THE SOUTHWEST TOWER OF THE GREAT TRANSEPT
Anno a[b incarnatione Domini] millesi{mo centesimo? consecra]tum est foratorium hoc cum altari in?] onore et [. . . Ihi] Xpi et Sce Ge[netricis Dei Mariae (?) et specjialit{er sub inuocatione? sancti Gab]rihelis [et sancti Raphaelis? archa]ngelor[um domini? et sa}nc[ti La]urencii et Sancti [Maxi- mi? omniuJm sanctorumque consecra{uit dominJus Petrus Panpilonensis episcopus 11 Idus marcii: et posuit ibi in altari reliquias Beati Laurentii leuite et martir et Sancti Maximi.
The text is inscribed in red paint on a coat of stucco applied over the original stucco interior, in letters of the period. Pedro de Roda was another Cluniac building bishop. He introduced the rule of St Augustine for his canons at Pamplona and reconstructed the cathe- dral. Small as the Chapel of St Gabriel is, it is of inescapable im- portance in dating the structure. A subsidiary feature of the great transept, placed as it is in the upper part of the building, it was doubtless built after the body of the transept was completed. Its sculptured capitals show by their cutting that they were carved when set into place. Some of them resemble the capitals of the lower part of the transept and the carvings farther east, including those of the minor transept and the ambulatory capitals themselves. The scale is smaller, but the beautiful composition and the acanthus mollis of the leafage are the same. Others of the chapel capitals fore- cast the carvings of the octagonal transeptal tower, which were made well along in the twelfth century. The chapel is at mid-point in the plan of the church, and it stands midway chronologically and sculpturally, as already shown in the Bulletin Monumental (loc. cit.).
Had the date in the inscription been preserved, the difficulties of the archaeologist would be less, but fortunately we have an ante quem in the death of the bishop in 1115 at Toulouse. The dedication is not mentioned elsewhere and there seems to be no record of a re- dedication in 1130-31-32. The tower and its chapel are thus seen to be anterior to, and undisturbed by, the disaster of 1125. They are probably as much as twenty years earlier for, as will appear, the es- sential parts of the church were measurably complete by 1113, in-
3 in le ill od | 5. c- y. he oh he or ed ed ni- obi ; bi, m, vi nd the ted sad of De-
84 Mediaeval Academy Excavations at Cluny, V
cluding two hundred and sixty feet of nave lying to the west of the great transept, while, as has been indicated above, there is reason to believe that the work extended but little to the west of the minor transept in 1095. Obviously the great transept was undertaken next; but it is likely that there was an interruption, for the axis of the great transept is not properly located in relation to the axis of the construction farther east. The excavations and measurements of 1929 contribute this new fact.
The main fabric of the south great transept as we have it, how- ever, is all of a piece. The writer, after a close study of the edifice in the course of several years, during which he has had occasion to examine the building repeatedly and thoroughly, in all its parts, feels no hesitation whatever in affirming categorically that there is no sign of an interruption in the main fabric below the Chapel of St Gabriel in the square tower, and the lowest stage of the octagonal tower. This latter tower was added in the twelfth century, as is clear from the cutting of the stones: they are not boldly worked with the hatchet like the older portions of the church, but with the pick in the manner already current when the Romanesque facade was constructed, early in the new century. The open stage of the square tower, with its curved roof, dates from the seventeenth century; the Chapel of St Martial, built in the fourteenth century, partly ruined and rebuilt in the nineteenth, is an obvious addition; and so are various less important modern works, including a sacristy which was erroneously described in SPECULUM, Iv (1929), 291, as a Ro- manesque construction. None of the new work is so placed as to put the continuity of the original work in doubt.
In addition we have the evidence of the cutting of the stones and the exterior design of the windows of the two southern bays of the transept. The hatchet work, the impost mouldings profiled only on the inner face, and the stuccoed arch-soffits, all characteristic of the eleventh century, are out of place after 1125. We may thus be cer- tain that the transept represents the eleventh-century design of the church. Since we know from old engravings that the design of the nave was similar, it is clear that the reconstruction of 1125-30 did not change the character of the building.
ti t!
Mediaeval Academy Excavations at Cluny, V