Illustrations of the History of Medieval Thought and Learning/APPENDIX

by Reginald Lane Poole

APPENDIX.

I. NOTE ON THE ORIGIN OF THE LEGEND RESPECTING JOHN SCOTUS’S TRAVELS IN GREECE.

It has been constantly repeated, as an old story to which modern critics cannot be expected to give credence, that John Scotus made a journey into Greece, and derived thence a part of the materials of his extraordinary learning. The story, however, is itself of entirely recent origin, and rests exclusively upon the authority of bishop Bale. His words are :

Ioannes Erigena. Brytannus natione, in Menevia Demetarum urbe, sen ad tanum JJavidis, ex patricio genitore natus, a quibusdam scriptoribus philosophus, ab aliis vero, sed extra lineam, Scotus cognominatur. Duni Anglos Daci crudeles bellis ac rapinis molestarent, et omnia illic essent tumultibus plena, longam ipse peregrinationem Athenas usque suscepit, annosque quam plures literis Graecis, Chaldaicis, et Arabicis insudavit. Oninia illic invisit philosophorum loca ac studia, imo et ipsum oraculum solis quod Aesculapius sibi olini construxerat. In quo, abstemio cuidam humilimus servivit ut sub illo abdita sciret philosophiae secreta. Inveniens tandem quod longo quaesierat labore, in Italiam et Galliam est reversus. The source of this passage is manifestly the following chapter in the Secretum Secretorum, otherwise known as the Liber Moralium de Regimine Principum, and vulgarly ascribed to Aristotle. I quote from the manuscript in the library of Corpus Christi College, Oxford, cod. cxlix. f. 4, adding in the margin a collation of the small Paris edition of 1520, fol. v.

Iohannes qui transtulit c librum istum films Patricii, linguarum interpretator peritissimus et fidelissimus, inquit, Non Clstumllbrum pro et

APPEND, i. reliqiii locum cl nee templum, in quibus philosophi e consue- d ncque. vcrunt componere et f reponere sua opera s secreta h quod non J deponere. visitavi ; nee aliquem peritissimum quern credidi k habere ali- >" quef e quam noticiam de scripturis l philosophicis quern non " exqui- k aHquem 11 " ^ V1 quousque veni ad oraculum solis, quod " construxerat I phydcb Esculapides " pro se. In quo inveni quemdam virum soli- " exquisiverini. tarium abstinentem, i studentem in philosophia peritissimum. u construxit. n petses. M ingenio excellentissimum. cui me humiliavi in quantum potui, <i ing. exc. de- servivi 1- dihgenter, et supplicavi devote ut mini ostenderet rW. et. secreta scripta illius oraculi : qui 8 mihi libenter tradidit. i Et tdfelt. inter "cetera x desideratum opus inveni, propter quod ad xcpusdesid. y ilium locum iveram, et tempore longissimo z laboraveram.

• iaboravi lum Q uo na bit il ad propria cum gaudio remeavi. Inde referens

a cum gaudio i> gracias multis modis creatori, et ad peticionem regis illus- grates raui- trissimi laboravi : c studens inter lin., vel studiisl et transtuli studui. primo ipsum de lingua Greca in a Caldeam, e et de nac in pro et de hac. Arabicam. In primis f igitur, sicut inveni in isto codice, transtuli librum g peritissimum Aristotelis, in quo h libro respondetur ad peticionem regis Alexandri sub hac forma. reg. Alex. pet.

I have been directed to this passage by a remark of Anthony à Wood that the said John, whether Scotus, or Erigena, or Patricias (for by all those names he is quarto. written by authors), was one of great learning in his time, and much respected by kings for his parts. Koger Bacon, a great critic in authors, gives him by the name of Patricius, the character of a most skilful and faithful interpreter of the tongues, and to whose memory we are indebted for some true copies of certain works of Aristotle. Wood then translates from the Corpus manuscript the passage, which I have given above in the original, and which he supposed to be by Koger Bacon because the glosses in the volume are ascribed to him. The extract however is taken not from the glosses, but from the text itself; a text which might as well have been quoted from one of the printed editions, so that Roger Bacon s name should not have been introduced into the matter at all. As it is, Bacon has been treated for centuries as the author of a fiction of which, so far as I can trace, the proper credit belongs to Bale. 1 Fabricius in fact long ago found this out : Baleus hanc versionem libri de regimine principum male tribuit loanni Scoto Erigenae;^[1] the real John was a Spaniard. quarto -

II. EXCURSUS ON THE LATER HISTORY OF JOHN SCOTUS.

The statement that John Scotus retired into England after the death of Charles the Bald has been the subject of much discussion, and, as usually happens, the dispute has been complicated by a good deal of what is no real evidence, and by much confusion of the real and the false. The following extracts will put the reader in possession of the materials on which to form an opinion with respect to at least an important section of the enquiry.^[2]

1. Bishop Asser of Sherborne says that king Alfred

legates ultra mare ad Galliam magistros acquirere direxit, indeque " advocavit Grimbaldum sacerdotem et monachum, venerabilem videlicet virum, cantatorem optimum, et omni modo ecclesiasticis disciplinis et in divina scriptura erudi- tissimum, et omnibus bonis moribus ornatum ; lohannem quoque aeque presbyterum et monachum, acerrimi ingenii virum, et in omnibus disciplinis literatoriae artis eruditissi- mum, et in multis aliis artibus artificiosum. ; quorum doctrina regis ingenium multum dilatatum est et eos magna potestate ditavit et honoravit.

This record stands between the years 884 and 886, but in a digression of a general character relating more or less to Alfred s whole reign.^[3] Florence of Worcester, in quoting the passage, placed it as early as 872, and the only fact that we can presume as to its real date is that it probably refers to the state of peace subsequent to the treaty of Wedmore in 878. Afterwards, under the date of 886, occurs the famous passage describing the quarrel that arose at Oxford between Grimbald and his companions who had come there with him, and the old scholastics of the town. It was natural to suppose that these companions included that John already mentioned ; and such is the inference drawn in the Hyde annals, a. 886, according to which, Lib. monast. anno secundo adventus sancti Grimbaldi in Angliam, el E V Edtvards, iucepfa est universitas Oxoniae, . . . legentibus . . . Grim- baldo and others, the list ending with in geometria et astro- nomia docente loanne monacho et collega sancti Grimbaldi. Since, however, the passage in Asser relating to Oxford is known to be a modern interpolation, and since the Book of Hyde is a production not earlier than Edward the Third s reign, the evidence on this head may be wisely ignored. It is only necessary to add that one certain witness to the connexion shown by the passage first quoted from Asser, remains in king Alfred s preface to his translation of saint Gregory s Pastoral Care, which he says he learned P of Plegmund my archbishop, and of Asser my bishop, and of Grimfold my mass-priest, and of John my mass-priest.

2. At a long interval from the mention of the arrival of the two scholars, and in what is <i regarded as a quite dis- timct section of his book, Asser relates, a. 887, Alfred s foundation of the monastery of Athelney, and r describes its first abbat :

Primitus lohannem presbyterum monachum, scilicet Eald- saxonem genere, abbatem constituit; dcinde ultramarines presbyteros quosdam et diaconos ; ex quibus, cum nee adhuc tantum numerum quantum vellet haberet, comparavit etiam quamplurimos eiusdem gentis Gallicae, ex quibus quosdam infantes in eodem monasterio edoceri imperavit et subsequent! tempore ad monachicum habitum sublevari.

Asser proceeds to relate the attempted murder of abbat John by the servants of two Gaulish monks in the house. They waylaid him in church, and fell upon him with swords so that he nearly died. In regard to this passage it may be argued from the specification scilicet Ealdsaxonem genere^[4] that the author is introducing a new person whom he wishes to distinguish from the John already mentioned; at any rate Asser s words do not necessarily identify John the Saxon with John the comrade of Grimbald. It is, however, commonly held that sthe latter inference has a s predommant probability. The two stories we rind repeated by Florence of Worcester without any attempt at com- binmg them.

3. Hitherto we have had no mention of John Scotus. It is evident that he may be the John whose name is associated with that of Grimbald ; but it is impossible that he be John the Saxon. To combine the three was first attempted in the spurious compilation, undoubtedly a monkish forgery, as it is described by u sir Thomas Hardy, which goes under the name of abbat Ingulf of Croyland. Its author invents a mode of reconciling the different nationalities by making John not an Old Saxon, but simply summoned from Saxony.

Hinc sanctum Grimbaldum, artis musicae peritissimum et in divmis scriptuns eruditissimum, evocatum e rrancia, suo novo monasterio quod Wintoniae construxerat praefecit in abbatem. Similiter de veteri Saxonia lohannem, cognomine foho Scotum, acerrimi ingenii philosophum, ad se alliciens, Ade- lingiae monasterii sui constituit praelatum. Ambo isti doctores literatissimi, sacerdotes gradu et professione monachi sanctissimi erant.

The forger has merely confused Asser by importing into his narrative the name of John Scotus, which he knew, evidently, from the story long before made popular by William of Malmesbury. y lib. ii. 122 pp. 189 sq. ed. T. D. Hardy.

z lib. v. 240 pp. 392 sq., ed. Hamilton.

4. This story is told by William in three separate works, in the ^[5] Gesta Regum, the ^[6] Gesta Pontificum, and in a letter addressed to his friend Peter. The second of these accounts also rëappears, nearly word for word, in what is known as the Second Chronicle of Simeon of Durham; but this has no claim to be regarded as an independent authority.^[7] Of William's three narratives, that contained in the epistle to Peter, which is entirely occupied with the subject of John Scotus, is the most complete, and I give it here as printed by Gale, e cod. Thuaneo ms., among the Testimonia prefixed to his edition of the De Divisione Naturae.^[8] From the point in the course of this letter, at which William's other works introduce the narrative about John Scotus and thenceforward run parallel with it, I give at the foot of the page a collation of them as well.

Petro suo Willelmus suus divinae philosophiae participium.

Fraternae dilectioni morem, frater amantisime, geris, quod me tam ardua consultatione dignaris. Est enim praesumtio caritatis, quod me tanto muneri non imparem arbitraris. Praecipis enim ut mittam in litteras, unde Ioannes Scottus oriundus, ubi defunctus fuerit, quem auctorem libri, qui περὶ φύσεων vocatur, communis opinio consentit: simulque, quia de libro illo sinister rumor aspersit, brevi scripto elucidem, quae potissimum fidei videantur adversari catholicae. Et primum quidem ut puto probe faciam si promte expediam, quia me talium rerum veritas non lateat: alterum vero, ut hominem orbi Latino merito scientiae notissimum, diuque vita et invidia defunctum, in ius vocem, altius est quam vires meae spirare audeant. Nam et ego sponte refugio summorum virorum

5 The passage is not reprinted in the edition of Simeon in the Monumenta historica Britannica: see vol. i. 684 note b. It may be read in Twysden's Historiae Anglicanae Scriptores decem 148 sq., 1652 folio [and in T. Arnold's edition of Simeon's Works, 2. 115-117; 1885]. On the character of the Second Chronicle see the preface to the Monumenta, p. 88, and Hardy's Descriptive Catalogue, 2. 174 sqq.

6 [It is also found in the Royal MS. append. 85 f. 25 b in the British Museum, which was written in the eleventh or twelfth century and is certainly not autograph, as is asserted in the index to Hamilton's edition of the Gesta pontificum, 531 b. In the first edition of this book I printed a collation of this manuscript, but the text has since been published from it by Stubbs in his preface to the Gesta regum, 1. pp. cxliii-cxlvi.] laboribus insidiari, quia, ut quidam ait, Improbe facit qui in alieno libra ingeniosus est. Quapropter pene fuit ut iussis tarn imperiosis essem contrarius, nisi iamdudum constitisset animo, quod vobis in omnibus deferrcm, ut parenti gratissimo, in his etiam quae onerarent frontem, quae essent pudoris mei periculo.

loannes igitur cognomento Scottus opinantes quod eius gentis fuerit indigena, erroris ipse arguit, qui se Heruligenam in titulo Hierarchiae inscribit. Fuit autem gens Herulorum quondam potentissima in Pannonia, quam a Longobardis pene deletam eorundem prodit historia. Hie,^[9] relicta patria, Franciam ad Carolum Calvum venit, a quo magna dignatione susceptus, familiarium partium habebatur; transigebatque cum eo (ut alias dixi^[10]) tarn seria quam ioca, individuusque comes tarn mensae quam^[11] cubiculi erat : nee^[12] unquam inter eos fuit dissidium, quia miraculo scientiae eius rex captus, adversus magistrum quamvis ira praeproperum, nee dicto insurgere vellet. Regis ergo rogatu Hierarchiam Dionysii de Graeco in Latinum de verbo verbum^[13] transtulit : quo fit ut vix intelligatur Latina^[14], quae volubilitate magis Graeca quam positione construitur nostra,^[15] composuit et^[16] librum quern Trepl </>Jcrewv /uitpicr/jiov,^[17] id est, de naturae divisione, titulavit propter quarundam perplexarum quaestionum solutionem^[18] bene utilem si tamen ignoscatur ei in quibusdam,^[19] quibus^[20] a Latinorum tramite deviavit, dum in Graecos nimium^[21] oculos intendit.^[22] Fuit multae lectionis et curiosae, acris sed inelegantis, ut dixi, ad interpretandum scientiae; quod eum (ut verbis Anastasii Romanae ecclesiae bibliothecarii loquar) non egisse aliam ob causam existimo, nisi quia, cum esset humilis spiritu, non praesumsit verbi proprietatena deserere, ne aliquo modo a sensus veritate decideret. Doctus ad invidiam, ut Graecorum pedissequus, qui multa quae non recipiant auros Latinae, libris suis asperserit: quae non ignorans quani invidiosa lectoribus essent, vel sub persona collocutoris sui, vel sub pallio Graecorum occulebat. Quapropter^[23] et haereticus putatus est, et scripsit^[24] contra eum quidam Florus. Sunt enim^[25] in libro περὶ φύσεων^[26] perplurima quae multorum aestimatione,^[27] a fide catholica^[28] exorbitare^[29] videantur. Huius opinionis^[30] cognoscitur fuisse^[31] Nicolaus papa, qui ait in epistola ad Carolum, Relatum est apostolatui nostro quod opus beati Dionysii Areopagitae, quod de divinis nominibus vel coelestibus ordinibus, Graeco descripsit eloquio, quidam^[32] vester Iohannes genere Scottus nuper in Latinum transtulerit; quod iuxta morem nobis mitti, et nostro debuit iudicio^[33] approbari, praesertim cum idem Ioannes, licet multae scientiae esse praedicetur, olim non sane sapere in quibusdam frequenti rumore diceretur.^[34] Itaque^[35] quod hactenus omissum est, vestra industria suppleat, et nobis praefatum opus sine ulla cunctatione mittat. Propter hanc ergo infamiam, ut^[36] credo, taeduit eum Franciae, venitque Angliam^[37] ad regem Aelfredum, cuius munificentia illectus, et magisterio eius, ut ex scriptis eius 36 intellexi, sublimis, Malmesburiae 37 resedit. Ubi post aliquot annos a pueris quos docebat, graphiis perfossus, 38 animam exuit tormento gravi et acerbo; ut dum iniquitas valida et manus infirma saepe frustraretur, et saepe impeteret, amaram mortem obiret. lacuit aliquandiu 39 in ecclesia ilia, 40 quae fuerat infandae caedis conscia ; sed ubi divinus favor multis noctibus super eum luccm indulsit igneam, admoniti monachi in maiorem eum 41 transtulerunt ecclesiam, et ad sinistram altaris posi- tum, 42 his praedicaverunt versibus martyrem : 43

Conditus hoc 44 tumulo, sanctus sophista loannes,
Qui ditatus erat vivens iam 45 dogmate miro,
Martyrio tandem meruit conscendere coelum,
Quo semper regnant cuncti per secula sancti. 46

Sed et Anastasius de insigni sanctitate adhuc viventem collaudat his verbis ad Carolum.

[Here follows an extract from Anastasius the librarian, to which William adds :]

Alternant ergo de laudibus eius et infamia diversa scripta, quamvis iampridem laudes praeponderaverint. Tantum arti- fici valuit eloquentia ut magisterio eius manus dederit omnis Gallia. Verum si qui maiorem audaciam anhelant, ut synodus quae tempore Nicolai papae secundi Turonis congregata cst, non in eum sed in scripta eius duriorem sententiam praecipi- tant. Sunt ergo haec fere quae controversial!! pariunt.

5. This is the account of John Scotus’s end which was received throughout the middle ages. The little that

36 G P regis. 44 G P Conditur hoc ; G R Clau- 37 G P Melduni. ditur in. 38 G P foratus. 45 G R and G P iam vivens. 39 G P here insert inhonora 46 The last two lines are in the sepultura. Gesta regum as follow : 40 G Pinbeati Laurentii ecclesia. Martyrio tandem Christi con- 41 For in maiorem eum, G P eum scendere regnum in maiorem. -fin the archetype Quo, meruit, regnant sancti per of G P, preserved at Magdalen secula cuncti. College, Oxford, cod. 172 p. 185, In the Gesta pontificum : eum is inserted above the line.] Martyrio tandem Christi con- 42 G P ponentes [in the Magdalen scendere regnum MS. corrected from positum]. Quo, meruit, regnant cuncti per 43 For his praedicaverunt ver- secula sancti. sibus martyrem, G P his martirium Here the two narratives end, so far eius versibus praedicaverunt. as the Scot is concerned. Vincent of Beauvais, to take but a single instance, says about him, is all derived, including the epitaph, through the channel of b Helinand, from William of Malmesbury. William has, in common with Asser, just three points, (a) that John was a learned man, (b) that he was invited from Gaul by king Alfred, and (c) that he taught in England ; in other words exactly what Asser relates about John the companion of Grimbald, with the exception of the notice that he was priest and monk : it has nothing correspond ing to what he says of John the Saxon. Apart from the question of nationality, the latter was made abbat of Athelney, and his life was attempted by the servants of two Gaulish brethren of the monastery ; whereas John the Scot, according to William of Malmesbury, went not to Athelney but to Malmesbury; he was not abbat, simply a teacher; was not wounded at the instigation of monks, but was actually killed by the boys whom he taught. The only point in common between the two is the name John.^[38]

6. With the epitaph quoted by William as commemorat ing this sanctus sophista loannes, we may connect a notice c Hist univ which is contained in a chronicle referred to by c du Boulay Pans. 2. 443. ag fa e Historia a Roberto Rege ad Mortem Philippi I :

In dialectica hi potentes extiterunt sophistae, loannes qui eandem artem sophisticam vocalem esse disseruit, Eobertus Parisiacensis, Eocelinus Compendiensis, Arnulphus Laudunen- sis. Hi loannis fuerunt sectatores qui etiam quamplure habuerunt auditores.

M. Haureau rejects the comparison with the Malmesbury inscription, but he is in the meshes of the old snare about John the Saxon. His caution in refusing to apply the inscription as a help to explain the Paris chronicle will be respected ; but when he urges on other grounds that the Johannes sophista of the latter is identical with John Scotus, we are entitled to use this conversely as evidence for the credibility of William of Malmesbury s account. M. Haureau s identification has since received powerful support from the arguments of Dr. von Prantl;^[39] and if their conclusion be accepted, it is surely reasonable to claim this John Scotus the Sophist as the same person with his contemporary John the Sophist, whose epitaph William records; especially when the latter, no doubt repeating an old tradition of the monastery, expressly identifies this sophist with the Scot.^[40] The extract in du Boulay is therefore a piece of evidence that converges with those in the preceding paragraphs to one centre. We may or may not believe all that William says, but this we may affirm, that his narrative is self-consistent and intelligible, and that it is incompatible with, and con tradictory to, the whole concoction with which the false Ingulf has entrapped our modern critics. ^[41]

7. Mabillon and others have objected that John Scotus could hardly have visited England so late as after the year 880. But there is no reason, because he is known to have gone to France before 847, to conclude that he must have been born before 815. We may fairly presume that the young Scot came to the Frankish court when he was between twenty and thirty : he can hardly have been born much later than 825, but he may have been born as early as 815. But even should we accept an earlier date for John s birth, it does not follow as a matter of course that f since, according to Asset s account, he must have , 1 Huber 117. gone to England as late as 884, he must have been called by Alfred at an age when one can look forward to little or no future activity as a teacher, and when he could hardly have had much inclination to change his country and enter upon new surroundings. Setting aside the fact that Asser’s notice, if indeed it refers to John Scotus, is not placed under any particular date, it is evident that one cannot assert the impossibility of a man s working power lasting until or beyond his seventieth year. At the same time there is no positive ground for excluding the alternative date for John Scotus s birth, which would make him fifty-three in 878 or fifty-nine in 884.

8. Another question arises about John s ecclesiastical position. Here we must note that William of Malmesbury makes no mention of him as anything but a plain teacher. It is true that Staudenmaier, whose conclusion on this head is repeated by the slater biographers, insists that William’s John was abbat ; but the only reason he can give is that the historian relates the destruction of John s tomb in connexion with Warin de Lire s sacrilegious treatment of past abbats of Malmesbury. The passage is as follows :

Huic [Turoldo] substitutus est Warinus de Lira monachus. ... Is, cum primum ad abbatiam venit, antecessorum facta parvipendens, tipo quodam et nausia sanctorum corporum ferebatur. Ossa denique sanctae memoriae Meildulfi et cete- rorum qui, olim ibi abbates posteaque in pluribus locis antis- tites, ob reverentiam patroni sui Aldhelmi se in loco tumulatum iri iussissent, quos antiquitas veneranda in duobus lapideis crateris ex utraque parte altaris, dispositis inter cuiusque ossa ligneis intervallis, reverenter statuerat ; haec, inquam, omnia pariter conglobata, velut acervum ruderum, velut reliquias vilium mancipiorum, ecclesiae foribus alienavit. Et ne quid impudentiae deesset, etiam sanctum lohannem Scottum, quern pene pari quo sanctum Aldhelmum veneratione monachi cole- bant, extulit. Hos igitur omnes in extreme angulo basilicae sancti Michahelis, quam ipse dilatari et exaltari iusserat, inconsiderate occuli lapidibusque praecludi praecepit.

Reading this extract carefully, it should appear that we have just as much right to infer that William is carefully distinguishing between John and the abbats, as that he intends to identify them. It was Ingulf who first made John Scotus an abbat.

Returning then to the John, the companion of Grimbald, in the narrative of Asser, we find him described as priest and monk. Now all we know about John Scotus s clerical position from contemporary evidence is negative. Prudentius of Troyes, indeed, ridicules him for setting himself up as a disputant in a grave controversy, being barbarum et nullis ecclesiasticae dignitatis gradibus insignitum. But it is plain that his not holding any rank in the church, which is all the words need mean, does not involve the consequence that John was not ordained. Abailard, for instance, had, in all probability, only minor orders until he was in middle life ; yet he afterwards was appointed abbat. It is no doubt the fact that John is never styled priest or monk by any of his opponents : nor does he ever describe himself as such, after the prevailing fashion, in his writings. But the latter circumstance, at least, has a very natural explanation : he desired to rank as a philosopher, not as a priest. This is indeed, as Dr. Renter observes, a salient characteristic of his position in the history of Christian thought ; and it would be readily accepted by his enemies as a confirmation of their judge ment that he was a heretic. We are not to expect that, they would signalise, if they were aware of, his priestly calling.

9. On the other hand, it is a mistake to infer from the title of martyr, as to which even William of Malmesbury, m in one of his accounts, expressed a doubt, an identification with another John Scotus, who held a place in the martyrologies, at least in England and France, until 1586, when I presume it was discovered that the philosopher was unqualified for the dignity.^[42] It is strange that Staudenmaier and others who repeat the statement not observed Mabillon's refutation of it. There is no doubt that the martyr who was ommemorated on the 14th of November was John Scotus, bishop of Mecklenburg, who was killed on that day in 1066.^[43]

10. Milman attempts to select from the various opinions with regard to John Scotus's retirement into England. He thinks (a) that John fled into England under the denunciation of the church and the pope, apparently following William of Malmesbury, here disregarding the long interval between John's participation in the Gottschalk controversy and the earliest possible date for his withdrawal from France; (b) 'he is said to have taken refuge in Alfred's new university of Oxford.' In a note we read that 'the account of his death is borrowed by Matthew of Westminster from that of a later John the Saxon, who was stabbed by some monks in a quarrel,' which statement is evidently taken directly from Guizot's Cour d'Histoire moderne, 3. 174 sq. (1829). 'The flight to England,' adds Milman, 'does not depend on the truth of that story.' The writer known as Matthew of Westminster however did not borrow his story about John Scotus's death from an account of a later John the Saxon, but took his matter directly from William of Malmesbury.^[44] Besides, we have already seen the entire dissimilarity of the stories about John Scotus and John the Saxon.

11. In conclusion, if Asser intends to distinguish John the Old Saxon, the abbat of Athelney, from John the companion of Grimbald, it is possible that the latter is John Scotus. William of Malmesbury may have drawn a fact or two from what is said about the latter, but his account is altogether irreconcileable with the notice of the Old Saxon. It is the combination of the two persons mentioned by Asser, derived from the spurious authority of Ingulf, that has misled the modern critics, and induced most of them to discredit the narrative of William of Malmesbury, as though it depended upon that late forgery. William's account may therefore be judged by itself, and accepted or rejected just as we may rate the historian's general credibility: there is no reason for excluding these particular passages from that respect which those scholars who know William best are ready to pay to his honest, conscientious labours.^[45]

III. Note on a supposed Theological Exposition by
Gerbert.

In Bernard's Catalogi librorum manuscnvtorum Anqliae et Hiberniae, the Bodleian manuscript 2406, now known as Bodl. 343, is described as containing at f. 170b an Expositio in Canticum Gereberti Papae in Spiritum sanclum. Oudin by mistake quotes the title as in Canticum Canticorum. The Histoire littéraire de la France says that Gerbert 'composa un Cantique sur le saint Esprit, qui avec son commentaire faisoit autrefois partie des manuscrits de Thomas Bodlei, sous le nombre 1406. [sic] 10,' and the writer speculates as to its date and contents. The manuscript itself, however, at the place indicated, contains, at the end of a volume of Anglo-Saxon homilies, a page filled up in a thirteenth-century hand with glosses upon a sequence for the feast of saint Michael the Archangel, which is known from the Sarum Missal, and of which the authorship is apparently claimed by the glossator for Gerbert,^[46] Whether also a portion of the glosses themselves is to be regarded as Gerbert's composition, I must leave undecided: certainly the introductory passage proclaims itself to be the work of a commentator. Hitherto the text of the glosses has only been published in M. Olleris's edition of Gerbert's works. The editor gives the following account of it—'On attribue à Gerbert un cantique sur le saint-Esprit, cantica de s. Spiritu, conservé dans la bibliothèque Bodléienne, et une prose ajoutée au canon de la messe en honneur des anges. . . M. H. O. Coxe n'a pas trouvé le cantique, et il a eu la complaisance de copier lui-même le commentaire suivant.' Since, however, M. Olleris could not identify the canticum nor print the glosses without a multitude of grammatical and other mistakes, I have transcribed the text afresh; and I have prefixed the sequence to which it refers, and with the punctuation unaltered, from the edition of the Sarum Missal published at Paris in 1555, folio:—

Ad celebres rex celice laudes cuncta.
Pangat nunc canora cater va symphonia.
Odas atque solvat concio tibi nostra.
Cum iam renovatur Michaelis inclyta valde festa.
Per quem letabunda perornatur machina mundi tota.
Novies distincta: pneumatum sunt agmina per te facta.
Sed cum vis facis hec flammea ceu rutilantia sydera.
Inter primeva sunt hec nam creata tua, cum simus nos ultima factura: sed imago tua.
Theologa categorizent symbola nobis hec ter tripartita: per privata officia.
Plebs angelica, phalanx et archangelica principans turma, virtus uranica, ac potestas almiphonia
Dominantia nunima, divinaque subsellia cherubin etherea ac seraphim ignicoma.
Vos o Michael celi satrapa, Gabrielque vera dans verbi nuncia.
Atque Raphael vite vernula: conferte nos inter paradisicolas.
Per quos patris cuncta complentur mandata que dat.
Eiusdem sophia: compar quoque pneuma: una permanent in usia.
Cui estis administrancia deo milia milium sacra.
Vices per bis quinas bis atque quingenta dena.

Centena millena assistunt in aula ad quani rex ovem cente-simam.
Verbigena drachmamque decimam vestra duxit super algamatha.
Vos per ethera nos per rura devia.
Pars electa armonie vota demus hyperlyrica cithara.
Ut post bella Michaelis inclyta.
Nostra deo sint accepta auream circa arani thymiamata.
Quo in celesti iam gloria.
Condecantemus alleluia.

Ad^[47] celebres, rez celice. In primo notandum quod hoc nomen canticum plures habet acceptiones. Dicitur enim canticum applausus qui fit ad laudem alicuius divitis, dicitur etiam canticum leticia de terrenis habita. Dicitur etiam canticum cantus quem fecerunt filii Israel quando rediere de servitute, id est, cantica canticorum. Dicitur etiam canticum omne tale gaudium quod fit de supercelestibus; et secundum hoc hie accipitur hoc nomen canticum. Sed querendum est, quis sit autor huius opusculi quod pre manibus habemus, que materia, que utilitas, que intencio, cui parti philosophie pertineat; ita dico si alicui pertineat. Autor huius operis fuit papa Girebertus, qui, cum mirabilis esset in omnibus actibus suis, precipue verborum et sententiarum erat perturbator. Materia est spiritus increatus et spiritus creatus. Spiritus increatus est ipse deus. Spiritus creati sunt, ut angeli et archangeli. Intentio est excitare animos auditorum ad laudes supercelestium. Utilitas est celestes laudes exercere. Sed videtur quod nulli parti filosophie pertineat, sed potius teologie. Theos idem est quod deus; logos, sermo. Theologia in duas dividitur species: in ypoteticam et apoteticam. Ypotetica est sermo trinitatis ad unitatem; et econtra apotetica est sermo de supercelestibus, ut de angelis et archangelis. Autor iste primo utitur apologo, quasi proemio tocius operis sequentis, captando benivolentiam ipsius creatoris, antequam incipiat opusculum suum; dicens, O rex. Sed quia hoc nomen convenit regibus nostris, adjunxit celice: et est composita diccio ab hoc nomine celum et hoc nomine cunctos, quasi cunctos celi. Cuncta caterva nostra, tain homines quam angeli. Clangay, id est quodam clanwre clamct ; vel PANGAT, id cst cantet. CUM NOSTRA SIMPHONIA, id est cum clara simphonia ; vel CANORA SIMPHONIA, id est sonora simphonia. Simphonia dicitur a sin, quod est con, et phonos, quod est sonus. Sic triumphus dicitur quasi trium phonos, quasi vox trium sonorum ; et potest dici ilia concordantia que est mentis et oris, vel ilia que est inter ipsos sonos plurium personarum. AD CELEBRES LAUDES, id est ad festivas laudes ; ATQUE NOSTRA CONTIO, idem est quod nostra catena. SOLVAT, id est quasi debitum reddat. ODAS, id est laudes ; unde dicitur in alio loco palinodas, id est duplices laudes, a palin, quod est duplex, et odas, quod est laudes. Deinde redit ad propositum. Sed notandum quod liec festivitas non fit de pugna que fuit inter Michaelem et draconem, sed de miraculo quod contigit in Gargano monte. CUM IAM FESTA MiCHAELis VALDE iNCLiTA. Cleos idem est quod gloria ; inde inclita, id est valde gloriosa, quia in ponitur ibi augmentative et non privative. Renovantur, id est awn.ua- tim quadam renovatione celebrantur. PER QUEM, id est Michae lem, PERORNANTUR, id est beiie ornantur, LETABUNDA; id est ilia festa sunt quadam leticia habundanter celebranda. TOTA MACHINA, id est cum omni illo quod coutinetur sub firmaineiito. MUNBI : Mundus dicitur microcosmos, a micros 57 quod est longus, et cosmus quod est mundus ; id est celestis mundus qui semper durat : vel microcosmus dicitur a micros, quod est brevis, et cosmus, mundus, id est minor mundus, id est ipse homo. Quia sicut mundus constat ex quatuor elementis, sic homo ex quatuor humoribus qui concordant quatuor elementis. Sanguis enim concordat aeri, quia calidus est et humidus sicut aer. Colera concordat igni, quia calida est et sieca sicut ignis. Flegma concordat aque, quia frigidus et humidus sicut aqua. Melancolia concordat terre, quia frigida et sicca sicut terra. Vel mundus dicitur a mundiori parte mundane machine, id est a firmamento. AGMINA, id est consortia. NEUPMATUM : hoc nomen neupma duplicem habet acceptionem et potest cognosci in scripcione. Quando sic scribitur, pneuma,^[48] per p et n, tune portendit iubilum, qui fit post antiphonam; qui iubilus non potest exprimi corde et ore, sed sono tantum : et fit ad designandum celeste gaudium, quod non potest corde nee ore, pre eius magnitudine, sod sono ct iutellectu tantum, demonstrari. Quando vero scribitur sic, mneuina, per m et n, vel sic per n tantum, neuma, tune significat spiritual, secundum quod hie accipitur. Mneumatum, id est spirituuin. Per te, id est deum. Facta : improprie utitur hie nomine facia, quia aliud est fieri et aliud est creari. Fieri enim est facere aliquid ex traduce,^[49] id est ex preiacenti materia ; creari est aliquid facere sine preiacenti materia, quia deus creavit angelos sine preiacenti materia et cotidie potest crearc sic cotidie novas animas. DISTINCTA, id est divisa. NOVIES, id est per novem, id quia novem sunt ordines angelorum. NAM CUM vis, id est quaiido vis. FACIS HEC FLAMMEA, id est accipere igneam formam quando nunciant hominibus : vel aliter, flammed, propter ardorem caritatis quern predicant hominibus. INTER PRIMEVA SUNT HEC : hie ostendit quod quodammodo antiquiores sunt angeli quam homines ; ut dicitur, In prindpio creavit deus celum et terrain. Per celum intellege celestia, per terrain, ter- restria ; et sic quemdam primatum habent angeli ante homines. CUM NOS SIMUS CREATA, id est procreata : et notandum quod aliud est creari et aliud procreari et aliud fieri : creari, ut superius dictum est, facere aliquid sine traduce; ut cum material! fit et forma, et cum forma fit et materia : procreari, id est procul creari, ut ex nuce lignum : et fieri pertinet ad ipsum hominem et proprie.^[50] Unde dicitur homo facitor. ULTIMA CREATA, id est procreata. SED YMAGO TUA : aliud est imago et aliud similitude. Ymago, quia nos imitamur deum in iusticia et sapientia et prudentia, quia ipse iustus est et ius- ticia ; sic et nos iusti per iusticiam, etc. : similitudo est in lineamentis corporis. CATHEGORIZENT, id est predicent. THEO- LOGA : quid sint theologa, superius dictum est. SIMBOLA : symbolum est communis proporcio vel comproporcio, ut in con- vivio ; et dicitur a sin, quod est con, et bolos, quid est proporcio. Et dicitur simbolum dominica oratio, scilicet Credo et Qui- cunque vult, ubi est colleccio plurium articulorum Christiane fidei; vel simbolum dicitur ministeria^[51] angelorum, quia sepe ea quo ministrant et alia significant. TER TRIPARTITA, id est per novenarium disposila. PER PRIVATA OFICIA, id est per propria ofcia. Notandum quod hoc nomen qfficim.i, quando scribitur per unum /, tune idem est quod servire ; et quando scribitur per duo /, tune idem est quod nocere, unde officit ei. PLEBS ANGELICA PHALANX ET ARCHANGELICA. Sed quia autor in sequentibus facit mentionem de gerarchia, ideo videndum est, quid sit gerarchia, et unde dicatur, et in quot species divi- datur. Gerarchia est legitimum nature rationalis dominium; et videndum est quid quodlibet membrum in hac descriptione positivum^[52] operetur. Dominium dicitur, quia in nullo loco est gerarchia nisi ubi sit dominium nature. Rationalis dicitur, quia bruta animalia habent dominium super alia, que non dicitur gerarchia, quia ibi non contingit, nee eis convenit. Legitimum dicitur, quia reges et huiusmodi habent potestatem super alios, et hie forsitan non habent secundum legem legitime. Gerarchia dicitur a gere, quod est sacer, et archos, quod est principatus sive dominium. Gerarchia in tres dividitur species ; in supercelestem, celestem, et subcelestem. Supercelestis est summe trinitatis ypostasica monarchia : ypostasion vel ypostasis idem est quod substancia. Celestis gerarchia est ordo angelicus, qui dividitur in novem ordines. Subcelestis gerarchia, id est apostolatus et archiepiscopatus et episcopatus, et huiusmodi. De supercelesti gerarchia fecit autor inferius mentionem quando dixit, PER VOS PATRIS CUNCTA COMPLENTUR MANDATA QUE DAT. De celesti gerarchia fecit mencionem quando dixit, vos PER ETHRA. De subcelesti gerarchia fecit mentionem quando dixit, vos PER RURA. Sed quia dixi superius quod celestis gerarchia dividitur in novem ordines, ideo videndum est, quid sit ordo, et qualiter dicatur ordo, et in quot species habeat dividi. Ordo angelicus, ut ait magister Johannes Scotus, est caractere theo- phanie simplicis el non imaginarie et reciproce umformis spintuum insignita multitude. Multitudo aponitur quia ordo angelicus non potest esse nisi ubi sit multitudo. Spirituum apponitur ad differencial!! hominum, quia sepe homines con- templantur ipsum creatorem per ipsas creaturas. Insignita caractere, id est quodammodo sigillata signo. Caracter idem est quod signum : et ideo caractere apponitur^[53] theophanie. Theophania dicitur a theos, quod est deus, et phanos, quod est visio sive contemplacio : undo theophania, id est visio dei. Simplicis apponitur ad differenciam composite contempla- tionis ; quia quedam contemplacio est simplex, quedam com- posita. Composita contemplacio etiam in duas dividitur species, in contemplacionem secundum sensum, et contempla- cionem secundum racionem. Secundum sensum fit contem placio, quando contemplamur deum creatorem per ipsas creaturas; scilicet per solem et per lunam et per stellas, et huiusmodi. Secundum racionem fit contemplacio, quando nos contemplamur coherenciam inter materiam et formam; unde scimus quod abunivit materiam et formam : et hec etiam contemplacio est composita, quia quedam compositio est materiei ad formam et forme ad materiam. Simplex contem placio est quo fit inter angelos, quia contemplantur deum prout est in maiestate sua, et non per aliquas creaturas. Non imaginarie apponitur, quia quedam contemplacio est imagin- aria, quedam non. Imaginaria est ilia contemplacio.

[The rest is wanting.]

IV. NOTE ON THE PRECURSORS OF NOMINALISM.

DR. VON PRANTL was the first to a explain how John Scotus could be reckoned as the founder of nominalism, and to define the limits within which this ascription could be justly claimed. M. Haureau had indeed previously interpreted the reference in du Boulay s chronicle b already quoted, in the same sense as Dr. von Prantl; but he was led to this conclusion by the help of a passage in the De Divisione Naturae which he misread in an inexplicable man- ner. John Scotus omits grammar and rhetoric from the class of strict sciences, because non de rerum natura tractare videntur, sed vel de regulis humanae vocis, &c. d M. Haureau understood this of dialectic and rhetoric, and thus actually inverted the real significance of John s position in respect de of the function of logic.

Some commentaries attributed to Rabanus Maurus discover so close an affinity to John Scotus s logical theory as to suggest that they are immediately derived from him.^[54] Dr. von Prantl, therefore, maintains that if genuine they can only be placed among Rabanus s latest productions, and thinks that they have been wrongly attributed to him. Dr. von Prantl s reasoning does not appear quite decisive, and the conflict asserted to exist between the views con tained in these glosses and in Rabanus’s other works is not perhaps so substantial as to be fatal to their common authorship. Nor is it impossible that the former are independent of John Scotus s influence.

The next symptom of a nominalistic tendency appears in certain glosses in a Paris manuscript (now numbered Fonds Latin 12,949), of which specimens are given by Cousin and M. Haureau. The latter, and before him h Charles de Remusat, claimed their authorship for Heric of Auxerre. Dr. von Prantl, on the contrary, i considers the major part to be by another, though contemporary, writer. But he is in error in saying that the codex itself gives a different author to one section of the glosses in dispute (those on the Isagoge). It is true that the line,

lepa hunc scripsi glossans utcunque libellum,

stands in f . 52 b, but lepa, which Cousin had noted with a query, is a later insertion, written over an erasure with room for about seven letters. This point was ascertained for me by the kindness of M. G. Saige.^[55]

The logical summary found in a metrical version in another Paris manuscript, to which k Dr. von Prantl refers, can hardly be admitted as material for the history of the time before Roscelin, until we are better informed about Append., its date. Cousin, who prints these hexameters, J describes 669. 47 them as of the tenth or eleventh century, and hints the possibility that they were dedicated to a man who died in 1107. We cannot, then, be sure that they are anterior to Berengar of Tours, or even to Roscelin.^[56]

V. EXCURSUS ON A SUPPOSED ANTICIPATION OF SAINT ANSELM.

1 . SAINT ANSELM has been generally regarded as the first writer in the course of the middle ages who put forth a formal argument in favour of the existence of a God. Dr. von Prantl, however, claims the priority for William abbat of Hirschau, and infers from the fact that William is known to have been in correspondence with Anselm, at a date anterior to the publication of his Monologium, that the latter derived from William the idea of framing the argument in question. Dr. von Prantl s hypothesis is contained in a paper printed in the first part of the Sitzungsberichle der koniglich Bayerischen Akademie der Wissenschaften zu Munchen for 1861 ; and his results on the particular point which I have stated are given in full in his m Gescliiclite der Logik im Abendlande. The two arguments, however, arc quite different, William s resting upon the design and orderly government of the universe, while Anselm s proceeds from the existence of relative good to that of an absolute Good ; a reasoning which he subsequently exchanged for the simple proof that the being of God is implied in our thought of him. Besides, it is clear that the link sought to be established is at best a plausible conjecture : we have no evidence that the two men corresponded on the subject. Still it would be a sufficiently interesting coincidence if we could show that the first attempt among Christians during the middle ages to prove the existence of a God suggested itself to these two contemporaries.

2. Dr. von Prantl thinks that the argument was derived from Constantine the Carthaginian, afterwards a monk of Monte Cassino, who died before the year 1072, and who had acquired it, together with the physical learning for which he was famous, during a scholar s life of near forty years in the Mohammadan east. It is certain that the argument from design appears in Arabian philosophy a century earlier,^[57] but there is no hint that it occurs in Constantine’s writings. William, it is added, was in Rome in 1075, a few years after Constantine s death, and may then have made the acquaintance with the latter s books, which his own productions show him to have turned to good account. We have, however, no information as to the date at which William himself wrote the treatise ; and an examination of the book will soon show us that it is really later by a couple of generations than its supposed date, and has only by a blunder been attributed to William of Hirschau.

3. The little volume of Philosophicarum el astronomicarum Institutiomim Guilielmi Hirsaugiensis dim Abbatis Libri tres, which was printed at Basle in 1531, quarto, is textually the same book with the Ueol Aiddgecav sive Elementorum aopp. 2 . 311- Philosophiae Libri IV, printed among the works of n Bede in the Basle edition of 1563, folio. This Ileql Aiddgean , however, although it is actually quoted as Bede s, and as a possible source of an opinion of Abailard, by so accom- oAbeiard 2 . plished a scholar as Charles de Remusat, has been generally 223 n., 307 n. recogn i se( j as t he work of William of Conches, certainly p vol. 2. 1230. since the publication of P Oudin s Commentarius de Scrip- o PP. 457-462. toribus Ecdesiasticis, and of the a twelfth volume of the Histoire litteraire de la France. As long ago too as 1838 Charles Jourdain pointed out that the work in question existed also in the twentieth volume of the Lyons Maxima Bibliotheca Patrum the title De Philosophia Mundi, and under the name of Honorius of Autun ;^[58] and neither Jourdain nor any other writer (previous to Dr. von Prantl) who had mastered the facts, with reference either to the HEQI Aiddeojv or to the De PTiilosophia Mundi, had any doubt that their, or rather its, authorship belonged to Wil liam of Conches. Nor is manuscript authority wanting : it is found with his name, to take a single example, in a manu script of r University college, Oxford, nr vi p. 389, under the title Philosophiae Compendium. The fact, however, that the contrary hypothesis is supported by a scholar so distinguished as Dr. von Prantl, even though he has failed to observe the identity to which I draw attention, seems to justify a renewed examination of the question, in order to ascertain whether the book already thrice obscured under the names of William of Hirschau, Bede, and Honorius of Autun, could by any possibility be by the first of the three. I shall cite the three recensions as Hirschau, Bede, and Honorius, premising that when I speak of identity I do not exclude divergences, often wide diver gences, extending not only to the interchange of unim portant words, inflexions, &c., but also to the order of words in a sentence, and even further; such, in fact, as one is prepared to find in works so carelessly reproduced as those of a medieval writer, not of the first rank, would naturally be.

4. In each edition the work bears a different title, and in Hirschau it is divided into three books, while the others have four. The manner in which quotations are introduced throws a curious light on the processes by which writings were adapted to different authors. The writer of the manuscript from which Bede is printed, intentionally effaced what occurred to him as incompatible with the age of the presumed author.^[59] He has, however, gone carelessly enough to work. After, for instance, changing s Constantini, which refers to the eleventh -century Carthaginian, into the plural * philoso- phorum, he has left secundum eum immediately after : and he suppresses the name u Johannitius, which indicates Honain ben-Isaac, a Jewish physician of the ninth century, while he leaves untouched the reference to this writer s medical treatise known as the Isagoge, possibly through an ignorant confusion with the work of Porphyry which exercised so signal an influence on the learning of the middle ages. Yet the citations of classical and sub- classical authors, some perhaps more obscure than Con- stantine, are as a rule correctly given. In one instance a reference has been obscured in Hirschau, apparently in the interest of his authorship ; it is suggested in Bede and is given fully in Honorius :

HONORIUS p. 999	A. BEDE 313.	HIRSCHAU 8.
Cuius expositionem si quisquaerat, inglos-sulis nostris super Platonem inveniat.	Cuius exponere, si quis quaerat in aliis nostris scrip- tis inveniet.	Cuics expositio alias est.

5. Of the three recensions of the treatise, Bede is by far the worst;^[60] as a rule it is inferior to Hirschau, while the latter is perhaps slightly inferior to Honorius. None of the three editions, however, is complete. Hirschau breaks off first, just after having introduced the subject of the soul, whereas Bede proceeds from that point for a page and a-half further and Honorius a few sentences further still, the additional matter consisting of nearly twelve chapters in Honorius. This continuation is partly occupied with y the soul, which, however, is only cursorily treated. The author then passes on to the ages of man and their characteristics, and thus arrives at the subject of a education . These last four chapters occur also in Hirschau, but at the beginning of the book, under the title of Aliquot philosophicae Sententiae. In the closing sentence of ’Bede,’ which also concludes the section prefixed to ’Hirschau,’ we read the following scheme of the order in which learning should proceed :

b Ordo vcro discendi talis est ut quia per eloquentiam omnis fit doctrma, prius instruamur^[61] in eloquentia cuius sunt ties partes. . . . Initiandi ergo sumus in grammatica, deinde in dialectica, postea in rhetorica. Quibus instruct! et ut armis muniti, ad studium philosophiae debemus accedere, cuius hie est ordo, ut prius in quadrivio, . . deinde in divina pagina, quippe ut per cognitionem creaturae ad cognitionem creatoris pervenianius.

This in reality opens a new division of the author s whole subject ; for, as Honorius continues, quoniam in omni doctrina grammatica praecedit, it is his design to treat of grammar and, we may presume, of the other studies in their order. Sed quoniam, he concludes, de propositis supra . . sectantes compendia diximus, ut animus lectoris alacrior ad caetera accedat, hie quartae partis longitudinem terminemus.

6. Hitherto I have assumed nothing with respect to the authorship of the work in question, although at the outset its absence from the list of William of Hirschau s works given by d Trittenheim, who had peculiar qualifications for knowing about the monastery of Hirschau, may seem to raise a presumption against its accuracy ; not to speak of the surprise with which we find that most orthodox abbat credited with a theology betraying only too evidently the influence of Abailard. I have limited myself to showing the identity of the three works, which had previously, as I thought, escaped detection. In this I have since learned that I was mistaken. The fact was pointed out by Dr. Valentin Rose in the Literarische Centralblatt so long ago as e June 16, 1861. The sequel was interesting, f Dr. von Prantl in a reply professed with remarkable courage his familiarity with the phenomenon of which Dr. Rose charitably supposed him to be ignorant. It was difficult to believe that a man could describe at length a treatise which he knew to be textually identical with another work printed under a different name, and purporting to belong to a different century, without a word of allusion to the latter.^[62] Dr. von Prantl added that he proposed to prove from further evidence that William of Conches had used the work of William of Hirschau. [In his second edition Dr. von Prantl suppressed the pages about William of Hirschau, and transplanted something from them into his account of William of Conches, pp. 127 sq.]

The blunder, however, has survived, and Dr. von Prantl s theory was treated seriously by professor Wagenmann in the & Goeltinqischen qelehren Anzeiqen for 1865 and by Dr.

VI. Excursus on the Writings of William of Conches.

1. The number and attribution of the works of William of Conches have always been a standing puzzle in medieval bibliography. It has already been stated that the book which forms the subject of the preceding excursus, and which has been confused among the editions of the vener able Bede, William of Hirschau, and Honorius of Autun, is now generally ascribed to William of Conches. But it will be best to assume nothing about it until we have gathered sufficient evidence to warrant a certain conclu sion. All William s productions hang so closely together that the proof that one of them is his involves all the rest : and if the following investigation goes over a good deal of ground which has already been covered by previous bibliographers, it does not in all points arrive at the same results as they have done.

2. The book that may serve as a foundation for our inquiry is the Dialogus de Subslantiis physicis ante annos ducentos conjedus a Wilhelmo aneponymo philosopho, pub lished in octavo at Strasburg in 1567.^[63] The editor, G. Gratarolo, a Basle physician, who discovered the book in Italy, apparently at Padua, took it (as appears from the title-page) to be a composition of the fourteenth century : the internal evidence, however, is decisive on this head. The dialogue is held between the author and a certain dux Normannorum et comes Andegavensium, a style by which only two persons could possibly be designated. One is Geoffrey the Fair, the husband of the empress Matilda, from the year 1143 or 1144, until his resignation of the duchy in 1150^[64]; the other is his son, our king Henry the Second, from the latter date until his accession to the English throne. Henry, however, is excluded^[65] by the i mention of the education of the duke s sons, since he only married in 1152. It may be observed that the belief that Henry was intended, combined with the mistaken inference from k John of Salisbury that William was about the year 1138 a teacher at Paris, plainly originated the fable which we read in i Oudin, that Henry the Second olim in curia regis Franciac enutritus et litteris in Parisiensi academia initiatus sub Guillelmo fuerat. The same passage which shows that Henry was not the interlocutor in the dialogue helps to fix the composition of the work within narrower limits. In te tamen, says William, et in filiis tuis aliquid spei consistit ; quos non, ut alii, ludo alearum sed studio literarum, tenera aetate imbuisti : cuius odorem diu servabunt. The dialogue was written therefore some time, probably some years, before Henry was of an age to be knighted, in 1149; and we shall not be far wrong

if we place it about the year 1145.

APPEND, vi. 3, The author describes himself at the opening of the sixth book : m P p. 210 S q. m g a au em q uae a magistris, dux serenissime, multotiens audivi, atque omnia quae recordatione usque ad meditationem memoriae commendavi, et ut firmius verba retinerem (quae irrevocabilia volant) stili officio designavi, et iam quae per viginti annos et eo amplius alios docui, adhuc vix plene et perfecte intelligo, vixque intellecta propriis et apertis verbis explicare valeo : et unde mihi tarn hebes ingenium, tarn rnodica memoria, tarn imperfecta eloquentia ? an quia in patria vervecum 76 crassoque sub acre Nordmanniae sum natus ? alios affirmare audio non solum minima, sed etiam maxima, quae nunquam a magistris audierunt, per se intellexisse, nihilque esse tarn inusitatum, tarn difficile, quod si sibi ostensum fuerit, statim non intelligant atque expedite alios doceant. The passage therefore tells us what William s native country was, -and we have only to add the concordant cf. Haurdau, testimony of n all the known manuscripts of the work, Htt. 246. e which bear any title, to identify the place as a matter of certainty with Conches; it tells us also the author s age, as having been a teacher since about 1120-1125, besides some other particulars about him to which we shall return hereafter. 4. Walter of Saint Victor in his polemic against the opinions of Abailard, Gilbert of La Porree, Peter Lombard, and Peter of Poitiers, written about the year 1180, expressly mentions, in his fourth book, William of Conches as having adopted the Epicurean doctrine of atoms : Quae forte Democritus cum Epicuro suo atomos vocal. Inde Willielmus. de Conchis ex atomorum, id est, minutissimonnn corporum, concretione fieri omnia. The passage occurs among the 78 The edition reads Vernecum Vervecum in patria crassoque sub for vervecum, as though it were a aero nasci. proper name : the reference, how- M. Haureau had the right reading ever, to Juvenal, Sat. x. 49, 50, is in his manuscript, and translates obvious, la patrie des belier.V p. 231. [It Summos posse viros et magna ex- is found also in the Arundel MS. empta daturos 377 f. 131 in the British Museum.] copious extracts from Walter given by du Boulay, and APPEND, vi. the reference is to the dialogue i. pp. 25 sqq. :

Sunt igitur in unoquoque corpore minima, quae simul mncta unum magnum constituunt. Haec a nobis dicuntur elementa.

The interlocutor here objects, Ut mihi videtur, in sen- tentiam Epicureorum furtim relaberis, qui dixerunt mundum constare ex atomis : to which the author replies,

Nulla est tarn falsa secta quae non habeat aliquid veri ad- mixtum; sed tamen illud admixtione cuiusdam falsi obfus- catur. In hoc vero quod dixerunt Epicurei, mundum constare ex atomis, vere dixerunt : sed in hoc quod dixerunt, illas atomos sine principio fuisse, et diversas, permagnum et magne volitasse, deinde in quatuor magna corpora coactas fuisse, fabula est.^[66]

5. In most manuscripts the work is called the Dragmaticon Philosophiae, dragmaticon being a synonym of dialogus. Ducange quotes a sentence describing it as a work conducted by means of question and answer, and <i Dr. Schaarschmidt, who does not profess to have seen the dialogue with which we are concerned, rightly corrects the title into Dramaticon. William, as it happens, himself explains the source of the title :

r Sed quia, similitude orationis mater est satietatis, satietas fastidii, nostram orationem dragmatice distinguemus. igitur, dux serenissime, interroga : philosophus sine nomine ad interrogata respondeat.

The published book was edited from a comparison of two manuscripts, one of which bore yet another title. The preface is headed Authoris Wilhelmi in suam Secun- dariam praefatio : nam hoc eius nomen fuit et haec libri inscriptlo, ut ex antique exemplar! constat. Possibly therefore the printed title De Substantiis physicis is an insertion of the editor. From the Secundaria we pass to a fourth title, namely, Secunda Philosophia, which appears in two manuscripts of the Staats- und Hof-Bibliothek at Munich. A fifth designation is found in a manuscript, nr xcv, of Corpus Christi college, Oxford, dating, accord ing to Coxe’s Catalogue, from the thirteenth century : Gulielmi de Conchis, alias Shelley, Universalis Philo- sophiae Libri III. per modum dialogi, &c. Sixthly, in one of the Digby manuscripts in the Bodleian library, nr cvii, the work is entitled Summa Magistri Willelmi de Conchis super naturalibus Quaestionibus et Responsionibus, &c. In the following pages I shall cite the book as the Dragmaticon.

6. We have now to inquire in what way it bears upon the other works of its author. Here its testimony is precise and unambiguous. It is a new edition of a former work entitled Philosophia, modified in concession, as would appear, to certain complaints on the score of heresy; and the passages thus altered or expunged are to be found in that work which in the preceding excursus was recognised under the different names of Bede, William of Hirschau, and Honorius of Autun. It is also known that s objections were raised to a work of William of Conches, entitled the Philosophia, which objections are substantially the same with those enumerated in the following paragraph of the Dragmaticon. I have inserted in the margin the corre sponding places in the Philosophia.

After announcing the subject of his treatise William proceeds :

’Est tamen de eadem materia libellus noster qui Philosophia inscribitur, quern in iuventute nostra imperfectum, utpote im- perfecti, composuimus ; in quo veris falsa admiscuimus, mul- taque necessaria praetermisimus. Est igitur nostrum consilium, quae in eo vera sunt hie apponere, falsa damnare, praetermissa supplere. Falsa vero ilia quae contra fidem. catholicam nobis in eo videntur esse, ante auspicium dictionis, nominatim dam nare (lignum duximus. Unde omnes qui ilium habent libellum rogamus quatenus ea nobiscum damnent et exterminent. Verba enim non faciunt haereticum sed defensio. u In illo diximus, in divinitate esse tria, potentiam, sapien- tiam, voluntatem : potentiam esse patrem, sapientiam esse filium, voluntatem spiritum sanctum. Sed quod dictum est de potentia quod sit pater, de voluntate quod sit spiritus sanctus, etsi possit quoquo modo defendi, tamen quia nee in evangelic nee in scripturis sanctorum patrum illud invenimus, propter illud apostoli damnamus, Prophanas novitates verbonim devita. De sapientia quod sit filius, non damnamus, cum apostolus dicat Christum dei virtutem et dei sapientiam.

In eodem conati sumus ostendere quomodo pater genuit filium, illud que quod dictum est, Generationem eius quis enna- rabit ? ideo esse dictum quod sit difficile, non quia impossible : hoc iterum damnamus, et aliis damnandum esse pronunciamus.

Cum in eodem de creatione primi hominis loqueremur diximus deum non ex Adam vel ex costa foeminam fecisse, ex limo qui coniunctus illi fuerat, ex quo viri corpus plasma - verat; ideoque translatitie esse dictum quod ex costa Adae facta sit foemina : hoc iterum damnamus damnandumque iudicamus, sanctae et divinae scripturae consentientes quae ait quod immisso sopore in Adam tulit deus unam costam de costis eius, ex qua materialiter corpus mulieris plasmavit. Haec sunt igitur quae in illo libro damnamus.

7. There is therefore no doubt that the early work of William of Conches to which reference is here made, is that same production which forms the subject of the preceding excursus and which, according to Dr. Wagenmann, actually bears the specific title of Philosophia Willihelmi Magistri in a Stuttgart manuscript.^[67] Of this the Dragmaticon is in fact a new edition, rewritten and cast in the form of a dialogue. The substantial agreement of the two has been already pointed out by professor Karl Werner in the Silzungsberichte der philosophisch-historischen Classe der kaiserlichen Akademie der Wissenschaften at Vienna. This point being established it remains to apply the evidence thus obtained to clear up the other disputed facts in William s bibliography. One of these may be cursorily mentioned before we attack more serious difficulties.

In the Philosophia i. 15, quoted above, there is a refer ence to glossulis nostris super Platonem, a reference which Cousin easily discovered in a Paris manuscript of which he gives extracts. Knowing however only the Honorius recension of the Philosophia, which is in its turn referred to as nostra Philosophia, in the glosses in question, Cousin supposed that the latter were by Honorius of Autun, because he failed to .observe the identity of the presumed Honorius with that printed as Bede; which b p. 669. latter b he rightly attributed to William of Conches.^[68] The glosses themselves are on the Timaeus, and abound in silent allusions to William s other works. Some of the definitions, those, for instance, of c philosophia and ingenium, occur verbally in the d Philosophia or the e Dragmaticon;^[69] but I am inclined to think that the quotation from the Dragmaticon is only apparent, and really comes from the Philosophia which f we have seen to be a fragment as we now have it. If this be so the Philosophia and the Timaeus glosses may have been written about the same time and naturally contain cross-references.

To this same early date are evidently also assignable a set of annotations on Boethius’s Consolation of Philosophy,^[70] of which extracts have been printed by Jourdain, which the editor claims to be the first real Commentary, as distinguished from formal glosses, with the partial exception of that of Bovo of Corvey, devoted to the favourite author of the middle ages.^[71] 8. Two other works of William of Conches, the Secunda Philosophia, and the Tertia Philosophia, are described in the twelfth volume of the Histoire litteraire de la France. They remain in manuscript at Paris ; but specimens, some chapters at length, and tables of contents, are printed by Cousin. The first, we are told, is a dialogue on anthropology between a master and a disciple ; the second, also a dialogue, is an abridgement of the author s system of cosmography, derived from the Philosophia. Had how- ever Cousin been acquainted with the Dragmaticon he would probably have suspected that this was the immediate source, and would have found that D. stands not for discipulus but for dux, the duke of Normandy to whom the work is dedicated. Moreover these works are not abridge ments at all. The one is a literal transcript of part of the Dragmaticon, the other is a set of disconnected extracts from it. The latter is taken from different parts of books ii. vi., and leaves off just before the point from which the former is transcribed. Of course it is impos sible to speak with absolute certainty from Cousin s speci mens, but the following details of collation suggest a sufficiently plain inference.^[72]

The Secunda Philosophia begins with the words Dicendum est, &c., which introduce the section on animals occupying the major part of the sixth book of the Dragmaticon. extracts which Cousin gives represent with trivial variants the identical text of the corresponding passages in the Dragmaticon, and the order of the thirty-five chapters is exactly the same. The two copies end with the same words.

The Tertia Philosophia contains ten chapters of which Cousin has printed the first. This is simply a set of extracts from the Dragmaticon. I take the sentences as they follow. Mundum . . . extra quern nihil est will be found in the Dragmaticon, ii. p. 41 ; Nota quod tempore Martii . . . moritur, in lib. iv. pp. 123 sq. ; Nota : dicit Constantinus . . . pessima, in lib. iv. pp. 127 sq.; Verbi gratia . . . iudica, in lib. iv. p. 128 ; Nota : in autumno . . . periclitantur homines, on the same page. Chapters ii. ix. from their headings correspond to passages in the fourth and fifth books of the Dragmaticon ; chapter x. to something near the beginning of the sixth. The extracts speak for them selves : the Tertia Philosophia is nothing more than a note-book of selections from the Dragmaticon.

Such are the ’valuable fragments’ from which later scholars have drawn. Beyond insignificant various readings they add nothing to what was already printed in a complete form in 1567.^[73] William s original works therefore (excluding his glosses) are now reduced to two : the early Philosophia and the corrected edition of the same, the Dragmaticon. Is there a third to be added ?

9. The literary historians speak of a Magna de Naturis Philosophia by William of Conches as having been printed in folio, without place or date, about the year 1474. ^[74] This is a statement which has grown up by several stages. Josias Simler in his Epitome of Gesner s Bibliotheca, published at Zurich in 1574, says on p. 254 " that William scripsit philosophiani universalem lib. i. De naturis inferio- rum, seu philosophiani primam lib. i. De superiorura naturis, seu philosophiam secundam lib. i. Sunt autem duo magna volumina, ante multos annos impressa.

Then n Possevinus spoke of a work by William super Opere sex Dierum, of which he had seen only the volume beginning with book xix. His description leaves no doubt that the work he mentions is the second volume of Vincent of Beauvais s Speculum naturale in the edition, s. l. aid a., presumed to have been printed at Strasburg in 1468^[75] or 1473 (not in that of Nuremberg, assigned to the year 1483, and also in folio). The first page of this volume begins book xix. (after the table of contents) with an extract from William of Conches, headed con spicuously : De opere sexte diei. Et primo de amtnalibus. Guillerinus de conchis. This is the very title which has been constantly repeated as William’s by the bibliograpliers, and which even M. Haureau Ponce sought to restore to the catalogue of William s writings.^[76] In 1722 Casimir Oudin connected the description given by Possevinus with the statement in the Epitome of Gesner.

Scripsit igitur Guillelraus de Conchis Magnam de naturis Philosophiam, desumptam ferme verbotenus ex Operibus veterum Ecclesise Patrum. sq., Leipzig It was a book of extracts systematically arranged. But Oudin, too, had only seen the second volume printed with out date or place about 1474. The same work manifestly is intended by J. A. Fabricius, when he says of William of Conches :

Prodiit etiam sub tempus nascentis typographiae Thilosophia eius maior de natuns creaturarum superiorum, sive super opere sex dicrum hbn. xxxin. duo bus maioribus m toho voluminibus rarissime obviis, excusisque sine anni nota locive.

The authors of the Histoire litteraire de la France complicated the matter by erroneously asserting that Fabricius spoke of the book as in three volumes and confused it with the work of Vincent of Beauvais. Fabricius said two volumes, of which the second is beyond doubt the second volume of the Speculum natitrale. The probable inference is that the first volume of which Possevinus, Fabricius, and the authors of the Histoire litteraire were unable to find a copy, was likewise the first volume of the Speculum.

10. These last writers state, with Oudin, that the book contained little original matter, being mainly compiled by means of extracts from the fathers. Nevertheless they regard it as the source from which (a) the Philosophia, (b) the Secunda Philosophia, and (c) the Tertia Philosophia, were successively abridged ; a statement which has been repeated Cousin and others. Even the accurate Haureau, who had the Dragmaticon before him, said in the first edition of Philosophie scolastique, that the Secunda and Tertia Philosophia paraissent avoir ete faits pour venir a la suite de celui que nous venons de nommer, the Magna de Naturis Philosophia; si, toutefois, he adds, ils n en for- ment pas une partie. x It has further been asserted that the great work was largely used by Vincent of Beauvais in his Speculum nattirale ; but all the extracts from William which I have met with in it are taken either from the Philosophia or the Dragmaticon.^[77] 11. The character of the supposed Magna de Naturis Philosophia, as described, is in itself such as to arouse suspicion. For in William s known writings we do not find very many patristic quotations. His authorities are Hip pocrates, Plato and Aristotle, Cicero, Pliny, Ptolemy, Galen, Solinus, Macrobius, Boethius, Constantino, etc.; he draws illustrations from Lucretius, Virgil, Horace, Ovid, Seneca, Juvenal. But engaged as he was in the pursuit of natural philosophy and natural history, he had small occasion to quote the fathers, and his references to them seem to be limited to Augustin, Ambrose, Gregory, and Bede. In fact he expressly declares his independence, as a philosopher, of the fathers. In eis, he says, quae adfidem catholicam vel ad morum institutionem pertinent, non est fas Bedae vel alicui alii sanctorum patrum (citra scripturae sacrae authori- tatem) contradicere : in eis tamen quae ad philosophiam pertinent, si in aliquo errant, licet diversum qffirmare. This statement occurs in the y Dragmaticon, a work which we have seen to be scrupulously modified in deference to orthodox objections. It is therefore the less likely that, even before his plain-spoken Philosophia, William should have written a great philosophical work chiefly constructed of select passages from the fathers. Besides, if such be the nature of this Magna Philosophia, how can it contain the material which he subsequently, ex hypothesi, abridged, so as to form the Philosophia as we know it ? The latter, as I believe on account of this assumed chronological arrangement, the authors of the Histoire litteraire designate the Philosophia minor, a title, however, which they do not assert to be found in any manuscript or edition of it.^[78] I believe further that the entire basis of their theory rests on a misunderstanding of a passage in John of Salisbury, on which I shall comment in the ensuing excursus.

12. I have spoken of the Magna de Naturis Philosophia on the authority of those who profess to have seen the book and who declare that it bears this title z in most of the manuscripts. But since writing this and the foregoing excursus I have had the advantage of reading M. Haureau’s admirable criticism contained in the eighth chapter of his Singularity historiques el litteraires,^[79] He there states positively that no such manuscript exists in France, nor to his knowledge elsewhere. Accordingly he conjectures that the bibliographers mistook some other book, published about the same time, for William of Conches s ; and he suggests that the book in question is the De Universo of William of Auvergne. The precise identification will not serve, but there can be little doubt as I think, a con fusion with the Speculum nalurale that some blunder of this kind originated the whole theory which, it has already appeared, is so difficult to reconcile with the known facts about William of Conches.

VII. Excursus on the Interpretation of a Place in John of Salisbury’s Metalogicus, i. 24 pp. 784 sq.

1. William of Conches has been generally regarded as a teacher who abandoned the thorough and honest system of the school of Chartres in order to compete with the shallower and more pretentious masters of his day. The vol. 12. 457- Histoire litteraire de la France illustrates this defection by the instance of his work, the Philosophia, which it supposes to be an abridgement of a previous book, the very existence of which the preceding excursus has shown to be more than doubtful. Ce qui l’engagea, we are told, de composer cet abrege, ce fut vraisemblement l’envie de se conformer, ou plutot la necessite ou il se trouva de ceder au torrent des philosophes de son temps, qui decrioient la prolixite de leurs predecesseurs, et se piquoient de donner toute la philo sophic en deux ans. Car il est certain par la temoignage de Jean de Sarisberi, qu apres avoir longtemps resisté à ces sophistes, il se laissa entrainer par leur exemple, pour ne pas voir deserter son ecole. The same statement involves also the character of William s colleague, Richard l’Eveque, and is accordingly repeated under his article in the b fourteenth volume of the Histoire. It has become the accepted view in regard to William, and is adopted, to give a single instance, in Ritter’s Geschickte der Christlichen Philosophic. It is therefore the more necessary to subject the hypothesis to a close examination.^[80] The part of it, however, concerning the sequence of William s works needs no refutation, since it is directly contradicted by his own statement that he wrote the Philosophia in his youth, many years before John of Salisbury came in contact with him.

2. John of Salisbury’s words are as follows :

Ad huius magistri [Bernardi Carnotensis] formam praeceptores mei in grammatica, Gulielmus de Conchis et Richardus cognomento episcopus, officio nunc archidiaconus Constantiensis, vita et conversation vir bonus, suos discipulos aliquandiu informaverunt. Sed postmodum, ex quo opinio veritati praeiudicium fecit et homines videri quam esse philo- sophi maluerunt, professoresque artium se totam philosophiam brevius quam triennio aut biennio transfusuros auditoribus pollicebantur, impetu multitudinis imperitae victi, cesserunt. Exinde autem minus temporis et diligentiae in grammaticae studio impensum est, etc.

The language is no doubt ambiguous, and everything hangs on the sense we give to cesserunt. We may under stand the passage, Once they taught well, but after a while they yielded to the rush of incompetent rivals and followed their example ; or equally legitimately, Once these worthy successors of Bernard handed on his tradition, but after a while, disgusted with the prevalent method of teaching, they withdrew from the field. The words will bear either rendering; but John of Salisbury’s other evidence about his masters, as well as the incontrovertible language of William of Conches own writings, can only be reconciled with the second alternative : the first is altogether excluded by the known facts about William and Richard.

3. Taking first the testimony to be drawn from John of Salisbury s writings, we find that Richard l’Eveque remained through life a valued correspondent of his, and was consulted by him on exactly those points of scholarship on which, if Richard s career were as is commonly supposed, John would be the least likely to trust him. William of Conches died before John had become conspicuous in the learned world, but John’s recollections of master are uniformly honourable. He couples William’s name with those of Gilbert of La Porree, Abailard, and others of his most respected teachers, just by virtue of William s steady hostility to the empty-headed crammers of his day. John also speaks of the jealousy which William and his friends excited in the latter; but of their yielding in consequence of it there is not a word.

4. It is precisely to these envious detractors that William constantly alludes in the prefaces to that Philosophia which, according to the Histoire litteraire, he condensed in deference to their opinion. The evidence of the prefaces to books i., ii., and iii. bears directly on the point ; that of the two former, which I quote, is especially pertinent :

Multos tamen nomen magistri sibi usurpantes, non solum hoc agere sed etiam aliis sic esse agendum iurantes, cognosci- mus, nihil quippe de philosophia scientes, aliquid se nescire confiteri erubescentes, sive imperitiae solatium quaerentes, ea quae nesciunt nullius utilitatis minus cautis praedicant.

Quamvis multos ornatum verborum quaerere, paucos veritatem scire [al. scientiae] cognoscamus, ninil tamen de multitudine sed de paucorum probitate gloriantes, soli veritati insudamus.

Another passage answers the allegation of the Histoire litteraire in a curiously exact manner. Speaking of the duties of a teacher, William savs : 1 Sed si amore scientiae ad docendum accesserit, nee propter invidiam doctrinam subtrahet ; nee ut aliquid extorqueat, veri- tatem cogmtam lugiet ; ncc si denciet multitude sociorum, desinet; sed ad instructionem sui et aliorum vigil et diligens erit.

These quotations, I repeat, are taken from a work which, we are asked to believe, was shortened in concession to the rage for short and easy methods.

5. At a considerably later date William wrote the Dragmaticon, and in this the protests against the fashionable tendency are if possible stronger than in the Philosophia. One ironical reference to the author s constitutional dulness and incapacity to understand things after long thought, which his pretentious rivals professed to grasp in a moment, has been k already quoted. 1 In another he complains of the way in which the teachers of his time have lost credit among their scholars. Both he says are in fault ; for to establish confidence one needs two things, knowledge and uprightness :

Quia igitur omnes fere contemporanei nostri sine his duobus ofncium docendi aggrediuntur, causa sunt quare sibi minus credatur. Discipuli enim culpa non carent, qui relicta Pytha- goricae doctrinae forma (qua constitutum erat discipulum septem annis audire et credere, octavo demum anno interro- gare), ex quo scholas intrant, antequam sedeant, et interrogant, imo (quod deterius est) iudicant ; unius vero anni spacio negli- genter studentes, totam sapientiam sibi cessisse putantes, arreptis ab ea panniculis, vento garrulitatis et superbiae pleni, pondere rei vacui abeunt : et cum a suis parentibus vel ab aliis audiuntur, in verbis eorum parum aut nihil utilitatis per- penditur; statimque quod hoc solum a magistris acceperint, creditur undo magistri authoritas minuitur.

6. The words of John of Salisbury, as I construe them, read precisely as an echo of what we now find to have been the consistent attitude towards learning and teaching maintained by William alike in his earliest and in his latest works. It is right to add that I was led to my interpretation of the passage in dispute, by a comparison of John of Salisbury’s different references to William of Conches and Richard l’Évêque, and before I had entered upon the examination of William’s own writings. It may be doubted whether the common view which I combat would ever have been suggested, far less accepted, had the historians of medieval literature taken the trouble to acquaint themselves personally with the books they describe.

VIII. Note on Abailard’s Masters.

The manuscript of Saint Emmeram’s, Ratisbon (now at Munich), from which Fez printed Abailard’s Scito te ipsum and Rheinwald more recently the same writer s m Sententiae contains a notice of his biography which, it seems to me, is worthy of attention. The character of the works in the volume is such as to mark it as proceeding from the inner circle of Abailard s disciples ; for the Scito teipsum had the reputation at least of being peculiarly esoteric, in fact, like the Sic et non, of shunning the light.^[81] The presumption therefore is that the biographical record which accompanies these pieces is based upon special sources of information. Unfortunately a part of it is so evidently apocryphal that it has discredited the remainder. It runs as follows :

Petrus, qui Abelardus, a plerisque Baiolardus, dicitur, natione Anglicus, primum grammaticae et dialecticae, hinc divinitati operani dedit. Sed cum esset inaestimandae subtilitatis, inauditae memoriae, capacitatis supra humanum modum, auditor aliquando magistri Roscii, coepit eum cum exfestucatione quadam sensuum illius audire. Attamen im peravit sibi ut per annum lectionibus ipsius interesset. Mox ergo socios habere, et Parisius palam dialecticae atque divini- tatis lectiones dare coepit ; et facile omnes Franciae magistros in brevi supervenit. Qui cum de Quadruvio nihil audisset, clam magistro Tirrico in quasdam mathematicas lectiones aures dabat, in quibus supra quam aestimaret obtentu diffi- cultatis intellectus resiliebat audientis. Cui semel afflicto et indignant! per iocum magister Tirricus ait, Quid cam s plenus nisi lardum baiare consuevit ? Baiare autem lingere est. Exinde Baiolardus appellari coepit. Quod nonien tanquam ex defectu quodam sibi impositum cum abdicaret, sub littera- tura non dissimili Habelardum se nominari fecit, quasi qui haberet artium apud se summam et adipem.

Taking these statements in order, we remark—

1. That the natione Anglicus, Britannus having been obviously changed into an apparent synonym, gives the impression of the writer being but remotely acquainted with Abailard’s history.

2. On the other hand, the order of his studies is cor rectly given. We have, it is true, no information about the time when Abailard learned grammar and it must be presumed that the writer merely conjectured that Abailard followed what was after all the natural and customary curriculum.

3. But the mention of Roscius (though the corrupt form in which the name is given may be considered to tell both ways) is of distinct importance. For a long time this passage was the only one, besides the notice of Otto of Freising, that spoke of Abailard s personal relations with Roscelin; and Otto s testimony was Pcommonlv discredited, especially because Abailard in his Historia Calamitatum altogether ignored the fact. So soon however as Abailard’s Dialectic was printed, it was found that he was in all probability the person referred to under the abbreviated style of magistri nostri Ros. The discovery in a Munich manuscript of a letter unquestionably addressed by Koscelm to his former pupil (though here the names are indicated only by initials), has finally decided the matter, and to this extent confirmed the evidence of the record here under consideration.

4. The next point, namely, that Abailard was unversed in the arts of the quadrivium is also of importance, since it is incidentally corroborated by Abailard s own statement that he was ignorant of mathematics : after quoting a geometrical argument from Boethius, he adds,

Cuius quidem solutionis, etsi multas ab arithmeticis solu- tiones audierim, nullam tamen a me praeferendam iudico, quia eius artis ignarum omnino me cognosce.

5. Then follows the story of his attendance upon the lectures of master Tirric. After what twe have said about Theodoric or Terric of Chartres, it is natural that we should be disposed to identify him with this teacher of mathe matics, especially since Tirric is found among the audience at Abailard s trial at Soissons. But what raises this conjecture to a higher degree of probability is the circumstance that the extracts which u M. Haureau has recently printed from an unpublished treatise by Theodoric, show an evident partiality ior mathematical illustrations. The account then of Abailard s connexion with Tirric suits exactly with what we know from other sources of these scholars attitude towards mathematics.

6. The concluding story about the origin of the name Abailard is of course a figment. Apart from its grotesqueness and intrinsic improbability (especially when we remember that, on the narrator s showing, Abailard must have adopted a new name after he had acquired his remark able reputation as a teacher), there is sufficient evidence that the name is not unique. A little before Peter Abailard’s birth, a son of Humphrey the Norman and nephew of Robert Wiscard received the name of Abaielardus.

7. Dismissing this legend then, we find that our document names two of Abailard’s teachers, one of whom (though the name is corrupted) points to an established fact, and the other to one inherently probable. The chronology however presents serious difficulties. There is no interval after Abailard entered upon the study of theology in which we can plausibly insert the lessons he had from Tirric; so that I incline to believe that Abailard made a short stay at Chartres during his first years of student life, after he left Roscelin and before he reached- possibly on his road to- Paris; or at the latest during the period for which, suffering from ill-health or the hostility of William of Champeaux, he retired from the neighbourhood of Paris. However this may be, I see no reason for doubt ing the truth of the bare fact that Abailard did enter upon a course of learning under Tirric.

IX. Note on the Second Preface to Gilbert of La Porrée’s Commentary on Boethius.

1 . John of Salisbury states that after the events of the council at Rheims Gilbert continued to suffer from the injury then done to him by those who sought to convict him of heresy, and took means to vindicate his position. Scripsit ergo posted contra illos alterum prologum in expositionem Boethii sui, in quo quosdam, videlicet emulos suos, assent sic hereticorum vitare nomina, ut tamen err ores eorum sequantur et doceant. The date of this new preface appears not only from the words of John just quoted, but also from the fact that according to John s account it was addressed to the capitula or articles of faith which were only produced by saint Bernard at Rheims. It therefore forms a sort of summing-up of the case from Gilbert s side, and was written for his own satisfaction at some time after the controversy had come to an end.

2. This preface seems to have disappeared, but an important fragment of it has been brought to light by professor Usener of Bonn, in the fifth volume of the a Jahrbucher fur protestantische Theologie for 1879. b Dr. Usener says, Each of the four commentaries has its introduction, and although that to the first treatise De Trinitate is more extensive than the following ones, it is not more general in its character but is concerned with discussions raised by Boethius text : this is the preface beginning, c Omnium quae rebus percipiendis suppeditant rationum. But, says Dr. Usener, in a Vatican manuscript (Lat. 560) of the thirteenth century we find further Item alius prologus, and this also appears in a manuscript of Saint Victor. It was written, a he thinks, for a second edition of Gilbert s Commentary, after the council of Paris and thus presum ably in preparation for that of Rheims. The hypothesis is no doubt possible, but it is curious that Dr. Usener should be unacquainted with John of Salisbury s account, with which it is natural to connect this new preface. It is more curious that the editor should not have observed that this very preface, only in a briefer form, is to be found in the very edition of Boethius which Dr. Usener had in his hands (that of Basle 1570), prefixed not to the Commentary but to the e treatise of Boethius itself. The preface is therefore not a discovery ; it is only an enlarged edition of that identical general preface, the supposed absence of which puzzled Dr. Usener.

3. The new part is however of sufficient interest to be transcribed here, especially because when printed in the midst of a mass of old matter its importance does not immediately attract attention. It is inserted, after the words scriptoribus recedamus, before the concluding sentence, exactly where we should expect such an addition to be made ; and it runs as follows :

Quamvis nos ab eis dissentire sarriant quidam fennii atque preconii, qui cum nichil didicerint, opinione sua nesciunt nihil, homines sine ratione philosophi, sine visione prophete, precep- tores impossibilium, indices occultorum, quorum mores plurimis notos describere nil nostra interest. Ipsi vero tanquam excussi propriis aliena negotia curant et obliti suorum satiras sati- rorum [sic] de ceteris animi ingenio et vite honestate preclaris multarum personarum fingunt comedias. Qui etiam in Deum blasphemi illos de ipso profitentur errores quorum nomina diffitentur. Nam, ut ita dicatur, hereticorum catholici in Sabellii, Donati, Pelagii, et aliorum huiusmodi pestilencium verba iurati, horum nomina (eo quod edictis publicis dampnata noscuntur) cum catholicis detestantur, ut cum blasphemiarum caussis sint iuste dampnabiles, blasphemorum detestatione putentur indempnes : sed quia non tarn res nominibus quam nornina rebus accommodat impositio, quibuscunque res con- veniunt, nomina non convenire non possunt. Quoniam vere sunt, recte vocantur, Sabelliani, Donatiste, Pelagiani, et huius- modi. Et bene quod novi heretic! nil afferunt novi, ut ad im- probandum adinventiones novas novis sit laborandum inventis. Antiqua sunt dogmata, olim per preclari et exercitati ingenii viros evidentissimis atque necessariis rationibus improbata, quibus eadem novissimis his rediviva temporibus possunt re- fellere, quicunque recte intelligentes virorum illorum scriptis lectitandis invigilant. Sed qui neque legunt neque lecturiunt, ideoque scientiarum elementa, si qua prioribus annis attendere consueverant, post longa desuetudine desciverunt aut etiam corruptis artibus a via veritatis exorbitaverunt, has omnino rationes ignoraverunt. Quorum si forte aliqui humano errore aut potestate aliqua presunt aut prominent dignitate, pre- cipiunt ut verum falsum et falsum verum, iterumque bonum malum et malum bonum esse credatur : et quod impudentissi- inuni est, ad sui magnificenciam quoslibet infames magnificant et magnificos infamant. Sed quia non tarn cognitores quam cogniti resident, sepe contingit ut rerum consequentibus can- cellatis cuiuspiam boni fame aliquid illorum favor detrahat et vituperatio addat. Quod nimirum attendentes, illorum male- dicta de nostris moribus et precepta de rebus contempnimus. Nam neque mores nostros convictu neque rerum proprietates disciplina noverunt.

Then follows the concluding sentence of the printed edition, whose text I retain, appending the two variants that occur in Dr. Usener s copy :

Quae 8 autem a nobis scripta sunt bene exercitatis lectoribus non modo rationibus firma, verum etiam scripturis autenticis adeo consona esse videntur ut nostra non tarn inventa quam h furta esse credantur.

4. The personal reference of the added passage is exactly in the same spirit as that answer which John of Salisbury v. supra, reports Gilbert to have given when Bernard suggested an interview. It is also a valuable specimen of the language which could be used about the saint by neither an insig nificant nor an irreligious section of his contemporaries. But the addition to the preface, although partly agreeing closely with what John of Salisbury says about the new preface, does not cover the whole ground which he describes. Either therefore the new preface itself is lost, or rather has been curtailed to its present dimensions, or else possibly John has mixed up with his account of it reminiscences of his conversations with Gilbert on the subject, reminiscences perhaps of his master s former lectures, or even his own independent vindication of Gilbert derived from a study of the Commentary on Boethius.

X. Note on Clarenbald of Arras.

Clarebaldus, archdeacon of Arras, is named in the continuation of k Henry of Ghent, just after Peter Lombard, as having written a commentary on the books of Boethius On the Trinity, in which he argued against certain opinions of Gilbert of La Porree, condemned Abailard, and favoured saint Bernard. In the ^Gallia Christiana he appears as holding the office of provost of the church of Arras in 1152 and 1153 ; and since his successor emerges in the year 1160, it is presumed that he died before that date. His com mentary should therefore offer valuable contemporary evidence in regard to the controversies spoken of in my m vol. 12. 445. sixth chapter; but the m Histoire litteraire de la France says it is non imprime et peut-etre perdu. It exists,^[82] however, among the manuscripts of Balliol college, Oxford, in the very same volume, cod. ccxcvi, which contains some of Abailard’s most treasured writings.^[83] The Commentary was written after August 1153, since it speaks of n iocunde recordacionis abbas Bernardus. We learn also from it that the author his name is here spelled Clarenbaldus was a disciple of Hugh of Saint Victor, and of Theodoric the Breton, no doubt the famous chancellor of Chartres.

Has causas mihi aliquantulum pertinaciter investiganti doctores mei venerabiles, Hugo videlicet de Sancto Victore et Theodericus Brito reddidcre. Magister vero Gillebertus Picta- vensis episcopus verbis perplexis hanc caitsam reddit. Que tametsi dispendiosa videri possunt, tamen in medium proferam, ne tarn clarum doctorem cum famosis doctoribus ascribere videar invidere.

He therefore writes his criticism on Gilbert with the object, in part, of showing that his judgement of him is not influenced by any grudge against including the illustrious doctor in the same class with the famous doctors first named ; so I understand the concluding words of the quotation. He charges Gilbert, as q so many others did, with an excessive obscurity of style : sq.

Exemplum huius lucidissime planitiei magister Gillebertus f. 204. Pictavensis episcopus multo verborum circuitu tenebrosam obscuritatem inducit, liberatque verbis reni frivolam involventi- bus, ut credatur, etc.

Clarenbald even finds fault with Gilbert s logic, speaking of him as sfalsum sibi in logica fingens, aut certe male "ibid. intelligens principium, quod est hoc, etc. In one place he describes some views of his as expressly heretical and as having been condemned at the council of Rheims :

Ex hoc loco episcopi Pictavensis error ortus esse videtur, ut tres personas numero differentes esse assereret. . . . Ergo nee numero tres persone inter se differunt. Quum vero in concilio Remensi sub Eugenio papa super aliis rebus liber eius reprehensus dampnatusque tam scolarium lectionibus quam claustralium ademptus est, et hie error, utpote heresibus eius aliis nullo modo preferendus, ibi commemoratus non est, commodum mihi visum est verba quibus hunc ipsum locum pertransire voluit, in medium revocare.

With respect to Abailard Clarenbald s language is still more hostile ; he accuses him of virtually resuscitating the opinions of Arius :

Eandem pene heresim Petrus Abailardus nostris diebus, longo sopore antiquatam, renovavit ; cum spiritu iactancie et impietatis plenus, divinitati ignominiam inferre, sibi gloriam conatus est pa rare.

References

1. Gale also, in the Testimonia prefixed to his edition of the De divisione naturae, lays the mistake to Bale s charge, but without detecting its source.
2. [Since this book was first pub- lished William of Malmesbury s Gesta rcgum has been reedited by bishop Stubbs, 1889, and Asser s Life of King Alfred by Mr. W. H. Stevenson, Oxford, 1904.]
3. [Bishop Stubbs, prof, to William of Malmesbury’s Gesta regum, 2 p. xlv., gives evidence to show that Grimbald came to England from Flanders not earlier than 892 ; but Mr. Stevenson, Asser 308 sq., points out that Grimbald was not an uncommon name at his monastery of Saint Bertin, so that it is not certain that the two persons are the same.]
4. Ealdsaxo means a Saxon of continental Saxony as distin- guished from a Saxon of England. Gregory the Second, when recom- mending saint Boniface to his future converts, addressed the letter universe populo provinciae Altsaxonum, Jaffe, Biblioth. Rer. Germ. 3. 81; and Asser himself elsewhere mentions regionem antiquorum Saxonum quod Saxori- ice dicitur Ealdseaxum, p. 484 A.
5. y
6. z
7. 5
8. 6
9. At this point the other narra- tives begin. The following is the text of the Gesta pontificum with which I collate that of the Gesta regum : Huius tempore venit An- gliam [G R Hoc tempore creditur fuisse] lohannes Scottus, vir perspicacis ingenii et inultae facun- diae, qui dudum relicta patria [G R dudum incrcpantibus undique bellorum fragoribus in] Frantiam ad .Karolum Calvum transierat. A quo magna, &c. The Gesta regum proceeds at once to the sentence beginning in the text of the Epistle with the words Regis ergo [G R cuius : G P Caroli ergo] rogatu.
10. G P omit ut alias dixi.
11. G P et mensae et.
12. The rest of this sentence is wanting in the Gesta pontificum, which contain instead the famous stories about the Scot and the sot, and the little fishes and the fat clerks.
13. G R and G P Dionysii Areopagitae in Latinum de Graeco, verbum e verbo.
14. G P add littera.
15. G R omit quo fit to noslra.
16. G R and G P etiam.
17. G P Perifision merimnoi.
18. G R propter perplexitalem necessariarum quaestiomim solvendam ; G P propler perplex itatem quarundam quaestionum solvendam.
19. G R aliquibus.
20. G R prefix in.
21. G R and G P acriter.
22. After intendit the Gesta regum go on directly with Succedentibus annis munificentia Elfredi allectus, venit Angliam, et apud monasterium nostrum a pueris quos docebat graphiis, ut fertur, perforatus, etiam martyr aestimatus est: quod sub ambiguo ad iniuriam sanctae animae non dixerim, cum celebrem eius memoriam sepulchrum in sinistro latere altaris et epitophii prodant versus, scabri quidem et moderni temporis lima carentes, sed ab antiquo non adeo deformes. The verses follow. The Gesta pontificum omit the passage Fuit multae to occulebat, but from that point agree closely with the text of the Epistola.
23. G P quare.
24. For et scripsit, G P scripsitque.
25. After enim G P insert revera.
26. G P perifision.
27. For multorum aestimatione, G P nisi diligenter discutiantur.
28. 26 G P catholicorum.
29. G P abhorrentia.
30. G P insert particeps.
31. G P fuisse cognoscitur.
32. So G P as quoted by Gale: Hamilton's edition by error has quidem.
33. G P iuditio debuit.
34. G P dicatur.
35. G P omit this sentence.
36. G P omit ut.
37. G P omit Angliam.
38. [Mr. Stevenson observes, intr. to Asser, p. cxii. n. 2, that bishop Stubbs has, by one of his rare lapses, confounded Malmesbury’s account of John the Scot with that of John the Old Saxon in the Life’ by Asser ; but he has not detected the source of this confusion in Ingulf.]
39. I have since read the objec- tions of Dr. Deutsch, Peter Aba- lard 100 n. 3, which, though un- doubtedly of weight, appear to me to depend too much upon consider- ations as to the character and con- tents of a chronicle which we know in fact only through du Boulay.
40. [See however Mr. Stevenson’s note to Asser, 335, where the sophist is identified with Jo- hannes se wisa, whose burial at Malmesbury seems to be recorded later than 1020.]
41. [Most of the foreign scholars who have discussed this subject have ignorantly treated Ingulf as a genuine authority: so Gfrorer 3. 938, and the biographers of John Scotus, Staudenmaier i. 120, 137, 140, Huber 115 sq., Christlieb 51. In the first edition of this book I dealt at some length with their various criticisms. ]
42. Thus in Andre du Saussay’s Marty rologium Gallicanum, the name (which is given as tempore Caroli Calvi) is relegated to the appendix, vol. 2. 122(>, Paris 1637, folio.
43. It is curious to notice that Trittenheim dichotomises the Scot. According to him, De Scriptoribus ecclesiasticis, 119 sq., ed. Cologne 1546, quarto, 'Iohannes dictus Erigena' translated the 'Hierarchiam et libros Dionysii' with commentaries, 'et quaedam alia.' 'Johannes Scotus,' on the other hand, p. 115, was a pupil of Bede and a comrade of Alcuin; to him is due the exposition of saint Matthew, 'one book' [sic] De divisione naturae, and another book, De officiis humanis; 'alia quoque multa composuit,' adds Trittenheim, 'quae ad notitiam meam non venerunt.
44. Why do Milman and Haureau, Histoire de la Philosophie scolastique 1. 151, and so many others, refer to the so-called Matthew for facts which he only states at second or third hand?
45. 'A steady attempt,' says Stubbs in his preface to the Gesta regum, 1 p. x, 'to realise the position of the man and the book has had, in the case of the present Editor, the result of greatly enhancing his appreciation of both.'
46. [My friend the late Dr. H. M. Bannister informs me that this prose appears in a service-book of Saint Martial's of Limoges written between 988 and 996 in the Bibliothèque nationale at Paris, MS. 1118, and in an Autun troper written between 996 and 1024 in the Bibliothèque de l'Arsenal, MS. 1169. It was widely current before 1050.]
47. The initial is not filled in.
48. Originally written mecrocos- mus a mecro ; but dotted for cor- rection, and with an i in each case above the e. 58 Cod. pncma.
49. Apparently the original read- ig was ratione, which has been corrected into traduce. M. Olleris has divisione, but this the manu script will not allow.
50. This word is miswritten in the MS.
51. Cod. mist ia.
52. Cod. p 9 ltu.
53. This word is repeated also after theophanie.
54. Extracts are printed by Victor Cousin, Ouvrages inedits d Abe- lard, intr., pp. xvii, Ixxviii, Ixxix, and app., pp. 613 sqq.
55. [Subsequent examination has shown that the word is not lepa but lepa, and it has been suggested that the letter.? are the beginning of a Greek name, possibly ICPAHA. See L. Traube, in Neues Archiv, 18 (1892) 105, and E. K. Rand, Johannes Scottus, p. 84; 1906.]
56. [Further evidence of a nominalistic tendency is found in an anonymous commentary on the Categories attributed to saint Augustin, which is preserved in a tenth-century manuscript at Vienna. See Prantl, 2. 44 sq., in the second edition.]
57. See the passage cited in the Sitzun^sberichte, ubi supra, p. 20, n 55 from Dieterici, Die Natur- anschauung und Naturphilosophie der Araber im zehnten Jahrhun- dert, p. 162 ; Berlin 1801.
58. Jourdain claims the discovery in the Notices et Extraits des Manuscrits, 20 (2) 43, n. J Histoire litteraire impartially de- scribes the same work under the head both of Honorius (vol. 12. 178 sq.) and of William of Conches. M. Haureau, Smgulantes nis- toriques etlitteraires 243, supposes that the original ascription of the work to Honorius by the editors of the Maxima Bibliotheca Pat rum was a mere guess : this is improb- able.
59. M. Haureau, Singularites 238, has not noticed this peculiarity, and charges the editors with in- advertence in admitting a work as Bede’s which contained refer- ences to later writers. As a matter of fact M. Haureau takes his quo- tations from Honorius.
60. In a few cases it contains good readings, as in p. 316, where com- mixtio and coniunctio stand in an inverted order from that in Hir- schau 18, thu? rendering Dr. von Prantl’s emendation, p. 15 n. 39, superfluous,
61. I correct from Hirschau.
62. The work is described under William of Hirschau, Gesch. der Logik 2. 83-85; and under William of Conches, 2. 127 sq.
63. This at least is the date that appears in the two copies of this very rare work that I have used, one in the Stadtbibliothek at Zurich and the other in the Bodleian library. It has been repeatedly given as 1566; see the Histoire litteraire de la France 12. 464, and Haureau, Singularites historiques et litter- aires 246.
64. [See C. H. Haskins, Norman institutions 130; 1918.]
65. This, I see, is observed by M. Haureau, Singularites, 232 sq., who also notices the source of the statement that Henry was Wil- liam s pupil at Paris; although I do not find that he disputes the story that John of Salisbury heard the latter there. Compare, how- ever, above, p. 181 n. 6.
66. Dr. Reuter verifies Walter’s citation in that work which is the subject of the foregoing excursus, and which, for reasons that will ap- pear immediately, I shall cite sim- ply as the Philosophia. He says, Geschichte der religiosen Auf- klarung in Mittelalter 2. 309 n. 28, that it occurs therein book i. ch. 21 (Honorius, pp. 999 & -1001 c); but in that passage there appears neither the reference to Epicurus nor the word atoms, while both are found in the dialogue. The authors of the Histoire litteraire de la France were unable to find the reference in any of William’s writings, vol. 12. 456.
67. The work also, according to M. Haureau, Singularites, 237 sq., bears the name of William of Conches in two Paris manuscripts; the titles added to the name, Tractatus Philosophiae and Philosophia are modern. M. Haureau, pp. 240 sq., takes the same argument as I have done from the Dragmaticon.
68. In his later edition, entitled Fragments philosophiques 2. 355, Cousin still only goes so far as to say that the glosses on the Timaeus pourralent bienetrede Guillaume . de Conches.
69. See other examples in Hau- reau, Singularites, 244.
70. At least they contain a precise declaration of a doctrine which William may be presumed to have withdrawn with his other impeached errors. See the quotation, above, p. 151, n. 11
71. The manuscript which contains the glosses on the Timaeus includes fragmentary commentary on Pri. cian, which M. Haureau, pp. 244 sq., conjectures is also by William.
72. M. Haureau in his Singularites still clings to the idea of these works being independent productions. I may, however, take leave to doubt whether this distinguished scholar had always the Dragmaticon itself before him. At least it is certain that every reference he makes to the Secunda Philosophia occurs, just as Cousin s do, in the Drag- maticon [e. g. ch. xviii. (Haureau, p. 252)= Dragm. p. 281; ch. xxx. (Haureau, p. 252 n. 2)= Dragm. p. 306] : and not in the fourth book of the Philosophia, as M. Haureau says (p. 241), nor anywhere else in that work. The substance may be there very possibly, though Cousin’s excerpts contain much that is definitely not there; but the form is that of a dialogue, and this fact alone decides the point, M. Haureau speaks (p. 247) of the Dragmaticon as borrowing from the Secunda Philosophia ; but when the smaller work is contained verbatim (within the limits of scrip- tural aberration) in the greater, we need not be long in deciding which is the original and which the extract. With regard to the Tertia Philosophia M. Haureau says little (p. 248), and does not seem to suspect that it is in fact derived from the Dragmaticon.
73. I have already stated, above 5, that the title Seeunda Philosophia is also borne by the complete Dragmaticon itself. The manuscripts thus entitled Dr. Reuter described as containing an entirely different work from Cousin’s Seeunda Philosophia, Geschichte der religiosen Aufklarung 2. 309 n. 30. What he quotes however certainly exists in the printed Dragmaticon, and I make no doubt that had Dr. Reuter read the manuscripts further he would have found all Cousin’s extracts there, as I have found them in the printed text. Moreover he misread Cousin, Ouvrages inedits d’Abelard, 669, and applied what the latter said of the Tertia Philosophia to the Seeunda. Here he was no doubt misled by M. Haureau, who speaks, p. 241, of part of the Seeunda Philosophia being borrowed directly from the Philosophia, book iv. The immediate source is incontestably the Dragmaticon, though the substance may often agree with that of the Philosophia. See preceding note.
74. The remainder of this excursus has been recast in the present edition.
75. It is attributed to Mentelin’s press under this date by Robert Proctor, Index to early printed Books, No. 255 ; 1898 quarto. Both volumes are in the Bodleian library, Auct. Q sub fen. 4, 5.
76. In correcting this mistake (which is repeated by cardinal Pitra, Spicileg. Solesm. 2. 188, Paris 1855 quarto), M. Haureau has fallen into a new one, in speak- ing, Singularites 236 sq., of the original as the Speculum historiale, in which what little is said about the sixth day of creation occurs in bk. ii. (misnumbered i.) ch. 38, and bk. xix. (opening with the history of Honorius) does not begin a volume.
77. For instance in book xxxii. 77, Vincent cites the latter as Guilhermus de Conchis without further specification, and then adds a quotation from the Philosophia as Ex libro de natura rerum.
78. William excuses the imper- fections of this book by the plea that, ’studiis docendi occupati, parum spacii ad scribendum ha bea- mus,’ lib. iii. praef. (Bed. 2. 330 ; Hon., p. 1010 B). This is scarcely the way in which an author would speak of abridgement.
79. M. Haureau’s essay, I have lately found, is in the main an en- largement of his article on William of Conches in the twenty-second volume of the Nouvelle Biographic générale, pp. 667-073; 1858.
80. The only writer I have found who interprets the passage of John of Salisbury as I do, is M. Leon Maitrc, Ecoles episcopates et mon- astiques 209; but he does not seem to be aware of the difference of opinion that has arisen on the point.
81. Sunt autcm, ut audio, adhuc alia eius opuscula quorum nomina sunt, Sic et non, Scito te ipsum, et alia quaedam, de quibus timeo ne, sicut monstruosi sunt nominis, sic etiam sint monstruosi dogmatis : sed, sicut dicunt, oderunt lucem nee etiam quaesita inveniuntur : Epist. Guill. de S. Theod. ad Gaufr. et Bern., (Bern. Opp. 1. 303 B, ep. cccxxvi. 4, ed. Ma billon). The Sententiae are coupled with the Scito te ipsum by Bernard, Ep. clxxxviii. 2, p, 181 E.
82. [R. Peiper mentions another manuscript, at Valenciennes, theol. 185 : pref. to Boot. Philos. Cons., 1871, p. 1.]
83. Among them the Commentary on the Epistle to the Romans which Cousin stated to be found in no known manuscript, although he had a portion of this very volume transcribed for him for his edition of another work of Abailard. [There is also a manuscript in the Vatican, Reg. Lat. 242 : see Denifle, Luther und Lutherbum i. 2. Quellenbelcge p. 49 (1905).]