STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world byJSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.istor.org/participate-istor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact support@jstor.org. EAKLY MEDIAEVAL COMMENTARIES ON TERENCE By E. K. Rand In 1893, Friedrich Schlee published what he called a despised and neglected set of scholia on Terence.' The manuscripts from which he excerpted them were with one exception familiar sources of the text of Terence: D ( = Laurent. XXXVIII, 24), G (= Vati- can. 1640), E (= Ricard. M xxx 528), C ( = Vatican. 3868). In these manuscripts the scholia are written in the margins or between the lines. Schlee colated besides a Munich manuscript of the eleventh century (CLM 14420 = M) in which the scholia do not accompany the text of Terence but, with the appropriate lemmata, form an independent commentary. Schlee's reason for bringing to light these tolerably unedifying contributions to the interpre- tation of Terence was that though the work as a whole was put together in the barbaric Middle Ages, the greater part of it, he believed, descended from antiquity,'' being connected with that ancient and important branch of the tradition represented by MSS D and A. The work was intended, he conjectured, as a school textbook, and as such had its place, seeing that the commentary of Donatus was far too elaborate for such a purpose. In publishing the scholia, Schlee proposed to winnow out their ancient portions, leaving the mediaeval chaff behind. "Recenti- ora scholia," he remarks (p. 48), "quae quidem certo cognosci possent omisi omnia ; si tamen pauca in tanta multitudine lateant, nemo id non excusabit." Accordingly, he divides the scholia into three classes: I. Quotations, or amplifications, from Servius and Priscian, from Donatus and Eugraphius, from Festus, Porphyrion, Isidore, and other authors. To these he gives the first place, because ' Scholia Terentianapraef., p. v : " quia vllia ab omnibus putabantur, neglectaadhuc iacuerunt ; " p. 49 : " primus hunc agrum patefeci, in quo nemo adhuc opus fecit, quia messis parum fructuosa videbatur." 'Ibid,, p. 50; "Sint sane aliquot recentioris aetatis, maior certe pars antiqua habenda est." [Classical Philology IV, October, 1909] 359 360 E. K Rand they are most obviously ancient in character. The number of such quotations varies in the different manuscripts. If we assume an original commentary in which some of them appeared, others must have been added sporadically in later copies. II. A Commentarius antiquior, consisting of brief explanations, numerous and minute. "Incredibile .... dictu est," remarks Schlee (p. 42), "quot verba, quot sententiae scholiastis interpreta- tione digna visa sint." These comments were originally headed by a life of Terence, he shows, and are themselves threefold in char- acter (p. 43): "aut verba explicant aut sententias circumeunt aut ad res pertinent." The last of these are few in number and not of much value, though the worst of them, he believes, are to be attributed not to the scholiast but to the scribe. The Latin of these comments Schlee sets in an extremely late period — "postremae Latinitatis" (p. 44). As sources he men- tions the Bembine scholia, Donatus, Eugraphius, Festus, Isi- dore, and others. Schlee believes that the scholia should be attributed to no one author; they "grew up" about the text of Terence. But the brief explanations which head the diff- erent scenes show a marked unity, and are attributed by Schlee to a single commentator, who wrote before the ninth century (p. 48). Wolfflin examined the Latinity of the prefatory ex- planations and sought to prove that they were written before the end of the empire and contained, furthermore, no peculiarities observable in Christian writers. Moreover, by emending what looks like a title on the first page of G, reported by Schlee as OOMMENTARII OOPEIVS, he discovered the author in the grammarian POMPEIVS of the fifth century.' But as Dziatzko clearly made out this title to be simply COMMENTATIO OPERIS there are no further reasons for associating Pompeius with this work.^ III. A Commentarius recentior, consisting of comments similar to those described under II, which include, however, longer pref- atory explanations and pay more attention to the development of ^Archiv.f. lat. lex. VIII (1893), 413 ft. ^JahrbilcherfUr Philol. XL (1894), p. 477. I am able to verify Dziatzko's state- ment from a grafflnS facsimile of G (the AndHa) in the possession of the Classical Department of Harvard College. Early Mediaeval Commentaries on Terence 361 the plot than to the exegesis of words. Such comments appear in the later folios of D and E, and come, Schlee believes, from some complete commentary of the eleventh century or later, of which various MSS in Rome, Munich, and Paris give evidence. To turn first to the Commentarius antiquior: Everyone will agree, I am convinced, with Schlee and Wolfflin, that the pref- atory explanations are by one author. If so, though it is conceivable that the original work consisted merely of scene intro- ductions, it is more probable that with due allowance for accre- tions, the bulk of the scholia belongs to the same author. The presence of substantially the same commentary in five or more MSS — for Schlee's list, as we shall see, may be enlarged — hardly argues for a piecemeal origin. Wofflin's stylistic analyses, too, indicate a unity in the body of the work, and though he distin- guishes an original pagan part and a Christian intermixture, one might understand how a later Christian writer could compose simple explanations for scenes in Terence without introducing phraseology peculiarly Christian. Schlee himself pointed to a clear connection between one of the longer comments and the introduc- tory vita (pp. 43, 94) ; scholia and vita, so far as this one instance can show, are of a piece.' There is nothing, in short, in the arguments thus far alleged to discredit the supposition of a complete commentary on Terence, composed by a Christian author, containing a preface, with a life of Terence, an introduction to the plays and the scenes, comments long and short on words and subject-matter. Further, as Schlee has pointed to the influence of Donatus, Paulus, Festus, Isidore, Eugraphius, and others, it is likely that some at least of the authors mentioned in Class I were used for the original work. Dziatzko, in his review of Schlee,^ speaks of the Co7n. ant. and Rec. as "aus sehr verschiedenartigen und -wertigen Bestandtheilen zusammengesetzte Kommentare mit einer gewissen einheitlichen Redaktion." As for the date of the Com. ant., the notes borrowed or adapted from Paul the Deacon's revision of Festus are signifi- 1 Dziatzko Jahrb. f. Philol., p. 475, states that the comment in question is not in M, and presumably only in D. But see below, pp. 362, 369. ^Deutsche Litteraturztg. (1894), p. 432. 362 E. K. Band cant,' in case they are an integral part of the commentary, which therefore could not have been put together before the last part of the eighth century. As I show below (pp. 373 f.), however, it is not safe to appeal to the evidence of these glosses. But waiving this last point we have evidence enough in the character of many of the comments that the work is of mediaeval origin. Schlee admits that in the process of excerpting, he may have allowed a few later bits to creep in; he certainly rejected much, for Dziatzko informs us^ that hardly a third of the material in M is retained for Com. ant. But how Schlee could include the following entries as evidence of antiquity is past comprehension. Eun. Didasc. (Schlee, p. 94) : " Quod Menander primnm com- posuerat fabulam Eunuchi in graeco sermone, deinde vero Teren- tius in latino sermone, iam dictum est et quod eius recitator fuit Calliopius." The scholiast imagines Calliopius a contemporary of Terence who recited his plays. It is Calliopius who pronounces the final "Plaudite," while "recensui" of the subscription is but another way of saying "recitavi" (Schlee, p. 94). Calliopius, too, delivers the prologues of the Hautontimoroumenos (p. 113) and of the Hecyra (p. 140) ; both passages inform us that Terence's faithful reciter was getting old. The Com. rec. (p. 165) has the same conception of Calliopius. Seeing that the successful play- wright needs three supports, a corrector, a reciter, a defender, Terence has them all, viz., "Laelium .... correctorem," "Cal- liopium declamatorem," "Scipionemdefensorem." Such blunders seem an impossibility for the age of Donatus or any period in the later Empire, but they are eminently characteristic of the scholar- ship of the early Middle Ages. Another curiosity may be found in the comment on the Didascalia of the Eunuchus (Schlee, p. 94) : " Claudi tihiis dua- bus dextris] quasi tympanum, genus musicum, dictum eo, quod unam alteri breviorem habeat fistulam ad similitudinem claudi. dextris dicit eo, quod tibiae avium, gruium videlicet, et ceterarum dextrae aptiores loco fistularum habentur et meliores." That Flaccus accompanied Terence's plays on a Pan's pipe, which was 1 Schlee, pp. 44, 76, and WSlfflin op. eit., p. 414. 'Jahrb., loc. cit., p. 469. Eahly Mediaeval Commentaeies on Terence 363 named from its appearance "the limping instrument," and which further w£is known as a "tympanum," strikes me as information not vouchsafed to antiquity. Nor had Aristoxenus known that the right shin-bone of the crane was preferred for a "tibia dextra." I will rest this part of my case on these influences, to which various additions will be made in the subsequent argument. We may well wonder what the remaining two-thirds of M contain if such comments as the above impressed Schlee as ancient, or, even, could inadvertently be included by him. Thus far I have argued in a general way that the Commentarius antiquior is a unit. At least it cannot be derived merely from an ancient set of scene explanations plus random accretions. Further analysis, however, reveals a distinctly twofold structure, though of an entirely different nature from that imagined by Wslfllin and Schlee.' The twofold nature of the following com- ments will be obvious at a glance:' Eun. 2. 3. 38 (Schlee, p. 101): " infeliciter] plus est infeli- citatis malum quam incommoditatis." — C G Br. "ista varo infelicitas est, quod mihi evenit, non incommodum malum."— DM. Phorm. 5. 1. 7 (Schlee, p. 137) : "in tutoforet] in securo esset, ne fame pjriclitaretur." — C G E F Br. "Ne meretricio operi serviret." — DM. Ad. Prol. 1 (Schlee, p. 149) "Postquam poeta] more suo in prologo malevolis responditpoetis,quieiusscripturam reprehendere nitebantur, et loquitur de se quasi de alio." — CEF (partly) Br. "quod in aliis prologis executus est, hoc etiam in isto, ut scilicet perfidis respondere posset." — D M. Ihid. 2. 2. 31: "labascit] victusest."— CGFBr.; "iam deficit animo." — DM. 7Wd4. 4. 18: "efferri] publicari."— FBr.; ("publicare?") — CG; "palam fieri."— DM. Ibid. 4. 7. 21: "tesseris] locus tabularum aleae. — C G E F Br. ; "tabulis.— DM. 'This fact has also been noted by Wessner Bursians Jahresbericht OXIII (1902), p. 189. See below, p. 368. 'I cite besides Schlee's MSS, the Halensis of Bruns (Br.) and the Ambrosianus (P). See below, pp. 364 ft. 364 E. K. Rand Ibid. 3. 2: Here we have variant scene explanations, which I will not take the space to quote. Neither version is of a Latinity so strict that one would attribute it to a pagan of the empire rather than to a Christian of the Middle Ages. It will be noticed in the above illustrations that C G E (with P and Br.) agree in presenting one version, D and M the other. Some twenty instances of this phenomenon occur in Schlee's edi- tion, and in about seventy instances, D and M agree in a comment which appears in only one, and generally in none, of the other manuscripts. All the manuscripts, doubtless, have suffered from accretions, and both D and M seem conflated here and there. Accordingly, to separate the two commentaries, we need full and exact reports of the MSS already treated. Schlee is purposely incomplete, and not always accurate, as Professor Minton Warren showed in his supplementary edition of the Phormio glosses in C* Yet even with the material published by Schlee it is possible to distinguish the two works, and to point out their different character- istics. Before attempting this task, I may call attention to one or two other sources of information with which we may reckon at once. The scholia in the Ambrosianus (F) had been made accessible in A PICTURE CAUIOPIUS MS (WitkourRrturts) X /Av: part by Angelo Mai,^ a work to which Schlee refers (p. 88), but now every scholar may read them for himself, in the facsimile 1 Harvard Studies in Class. Phil. XII (1901), pp. 125 ff. 2 M. Accii Plauti fragmenta inedita, item ad P. Terentium commentationes (Milan, 1815). Early Mediaeval Commentabies on Tebenoe 365 reproduction of F.' They are pretty clearly, like the text, in a late ninth- or early tenth-century hand, and thus give us an earlier source than any other MS save C. Moreover, seeing that F and E are doubtless members of the same branch of the tradition,^ and that E, a fairly late MS, has suffered extensively from accre- tions,^ F should help to establish the original condition of the scholia on the E F branch. Two other sources of information were at Schlee's disposal. Geppert* in 1852 published several mediaeval vitae of Terence, with the prose arguments for the different plays ; the value of this material will be apparent later on.'* More important still is the edition of Terence by Paul Jacob Bruns, in 1811, in which, to quote his title-page, "textum ad fidem codicis Halensis anti- quissimi criticis nondum cogniti edidit .... scholia a vulgatis diversa ex eodem codice descripsit." This work, ignored by Mai in his publication of the F scholia in 1815, seems to have escaped the attention of more recent scholars as well — possibly because of that suspicious phrase "ad fidem codicis antiquissimi." Umpfen- bach cites a reading or two from Halensis,* and Dziatzko' refers to the life of Terence edited from the MS by Bruns. No one, to my knowledge, has called attention to the remarkable character of the work, which is not a text of Terence, revised here and there with the help of the Halensis,' but is an almost exact reproduction of the manuscript itself. Bruns remarks in his preface (I, p. xxiii) : "Codicis textum deserere mihi religioni fuit, et fortasse erunt qui mihi vitio vertant, quod non talia menda, qualia occur- runt pag. 72 v. 24 queret pro quaeret .... miserae pro misere ' In the De Vries-Sijthofl series of facsimiles, Leiden, 1903, with preface by E. Bethe. ^Bethe (p. 49) believes that three illustrated MSS were brought to Prance, and that the history of the illustrations in the MSS has no connection with that of their text. I find nothing, however, in the valuable material amassed by him to discredit the stemma on p. 364, which. I believe, would coincide with Leo's views (Gotting. Gelehrt. Anzeig. 1903, pp. 997 ft.). ^E. g., various notes from Donatus. See Schlee, p. 69. *"Zur Geschichte der terentianischen Textkritik" in Jahrb., Supplementband XVIII(1852), pp. 32ff. 'Seebelow, pp. 380ff. « E. g., for the line following Eun. 5. 6. 9. ' Op. cit., p. 472. 8 As Geppert apparently imagined, op. dt., p. 29: "wie im codex Halensis, den Bruns verglich." 366 E. K Band . ... in animo pro animo .... una cum aliis paucis a me revera sublatis expunxerim. Sed accurate codicem repraesentare volui in hac editione." In the present state of Terentian criticism, these words need no apology. The editor had made accessible, with very few cor- rections, a not unimportant MS of Terence of the mixed class. Its text is decidedly nearer to 7 than E is, while it contains the Andria, the first part of the Eunuchus, and the end of the Phormio, which portions are lacking in F.' For our present purpose the Halensis is important in that Bruns edited with the text its scholia too.^ Moreover, he edited them after the only proper method. Schlee's attempt to sift ancient from mediaeval material proved, as we have seen, both disastrous and futile. As a complete edition of the scholia is at present impossible, next best is a complete edition of an important source. The publication of everything in M, with all its mon- strosities and superfluities, would have been of vastly more service to scholarship than illustrative selections compiled, in some cases inaccurately, from various MSS.' For the scholia, apart from their pedagogical value, have importance not only as portraying the culture of the time in which they were written* but as con- tributing further material for the criticism of Terence's text. Thus it is plain that the D G and F E branches have suffered from the incorporation of these very glosses. Schlee is aware of this point, but his list of instances (p. 35) needs revision. More- 1 1 am indebted for this estimate of the text of the Halensis to Mr. R. H. Webb of the Harvard Graduate School, who has studied the matter carefully in connection with a forthcoming dissertation. He refers for a similar opinion of the scrupulous methods of Bruns to Ritschl Opusc. III. 286. 2 This fact, neglected by Schlee, did not escape Sabbadini (Studi ital. difil. class. II, 29 ff.), who treated the scholia cursorily, however, as his chief interest was in Donatus. He well observes that the scholia show a general unity in character, betray- ing Hprofonda medievalitd (p. 30). He recognized that they were identical with the scholia published by Mai, and represented a complete set. Wessner likewise notes the importance of the Bruns scholia; see Jahresbericht, loc. cit., pp. 189 ft. 'For similar remarks see Dziatzko, Deutsche Litteraturzeitung, loc. cit., p. 432. He publishes there the exact contents of the first page in M. Schopen De Terent. et Donat. eius interp. (1821), p. 49, in remarking "cum istasglossulasvilesdescribendo etiam peiores fecerit Brunsius," apparently charges him with inaccuracy. I believe rather that Bruns preferred not to emend the errors of his MS. Given by Dziatzko, op. cit., pp. 470 ff. 'Jahrb., loc. cit., p. 472. 2E. g., pp. 49, 67. *Stud. ital., ete. V. 314. 6 " Der Terenzkommentardes Eugraphius," Rhein. Mus. LXII (1907), p. 223. See below, p. 374. I purposely postponed the reading of Wessner's article and his review in Bursians Jahresbericht CXLIII (1902), pp. 188 ff., until I had my own material in shape. It is significant that investigations starting from opposite points of view should terminate independently in similar results. Sabbadini, too, believes that vita III was not by Eugraphius, but concludes also, rather curiously, that it did not form part of a com- mentary; see Stud. ital. difll. class. V (1897), p. 314. Eaely Mediaeval Commentaeies on Terence 369 various remarks on comedy which are not inspired by Donatus, Euanthius, or Eiigraphius. The writer knows at first or second hand, Horace's Art of Poetry, which, as Wessner has pointed out, was not used by Eugraphius. From this work, from Priscian and Orosius, from the text of Terence itself, and, I am confident, from absolutely no. other source save his own perverse imagination, the author constructed an account of Terence and his comedies the like of which had never appeared before. We find here the conception of Calliopius as recitator which Schlee correctly rea- soned must have stood in the preface. We are also informed that whereas many deny that Terence wrote in meter, they are refuted on the authority of Priscian and Rufinus.' The meters used we next learn, are all varieties of iambic, the prologues being in iambic catalectic, which means that six feet are reckoned as three. At the end of the Andria [Com. Br. I. 146), the remark is made that the play had no title and that therefore nothing was known of its performance either among the Greeks or among the Romans; evidently the author did not have Donatus at hand. To compensate for this, he is able, by citing Priscian, to confound the higher critics of his day, who held that Calliopius composed the argumenta and the prologues of the plays. Reading on in the preface to the other plays, we find in that to the Eunuchus the astounding information about "clauditibiae" above mentioned (pp. 362 f.). The Megalensian games, he also states, were celebrated at Megale, a city in Greece, in honor of the Olympian Jupiter. At Rome these were given by the curule aediles, so called because they had charge of the "aedes curiales." Turning to the prologue, we find that the "miles gloriosus" (1. 31) is Terence himself, "nobilis comicus quia militavit in poetria.'" The introduction to the Adelphi adds the statement that the Greek performance of the play was encouraged by the consuls iFor some even of Priscian's contemporaries Terence's meter had lost all effect; see G. L. (Keil) III. 418. 8. 'In the Halensis the prologue with its scholia precede {Com. Br. 1. 149), then after EXPLICIT PBOIiOGUS comes the Didascalia followed by EXPOSITIO TITULI GUM ARGUMENTO. The Periocha itself does not appear. I will not venture to explain the relation of this fact to the condition of the text in the various MSS of Terence at this point ; it suggests that the MS for which the comment was written was either 7 or /*. 370 E. K. Kand (Br. II. 3), that the Greek name was "Synapothnes" and its author "Contesdifilus."' As for the Hecyra {Com. Br. II. 95), we are told of the two unsuccessful performances, the latter failing because of the interrupting "ludi funebres." The play was called Hecyra, as that is the name of a place not far from Athens. Coming to the Phormio, we find in the Halensis no preface at all. But there was one, doubtless, in the original commentary, since in various of Geppert's MSS^ which agree with the Halensis in the prose arguments to the other plays, there appears an argument for the Phormio similar in style to these others. Nothing like a conclusion or summary for the entire commentary appears at the end of the Phormio or of any other play. Passing now to the body of the work, we find nothing, I believe, in the terse and rather skilful explanations of the scenes that might not have been written by a scholar of the Carolingian period. Not all the scenes in the Halensis are provided with explanations. Is. this because the scholia are defective in this MS, or because the scribe did not feel the need of ubiquitous accuracy ? Watson' shows that at all of the nine places where the 7 MSS have new scenes, the explanations of the Com. ant. are found, whereas they occur in only three of the six new scenes peculiar to 8; again this is just what we should expect if the scholia were written for a mixed MS. The minor comments are intended chiefly to make the course of the action and the meaning of the words perfectly intelligible. As Wolfflin remarks,* the synonyms selected, which seem puerile to us, were intended to enlarge the school-boy's vocabulary ; they resemble somewhat the translations of words or phrases that figure in modern editions of classical authors. Stage directions appear, such as the frequent "secum loquitur," or "adverbium blandi- entis" {Com. Br. I. 305), "irascentis" (I. 138), "affectuose" (1. 112), "iocose" (II. 202). The schoolmaster insisted, too, on grammatical and rhetorical comment; "ironicos" ("ironice"),* "paranomasia" (I. 45), "eclipsis" (I. 186), "themesis" (I. 197), 1 See below pp. 384 S. 2 Paris 7904 s. XII ; 7914 s. XI ; 14755 (St. Victor 750) s. XII ; 16235 (Sorbonne 507) s. X. See op. eft., p. 42. 8 Loc. eit. ■> Op. cit, p. 415. ^Ibid., p. 418. Early Mediaeval Commentabies on Tekence 371 "litotes" (II. 222), "asyndeton" (II. 250),"hypallage" (II. 34), "ephexegesis" (II. 58), are some of the terms that appear.' The author had evidently studied the ancient grammarians, as the references to Priscian in the vita suggest. Longer comments on the subject-matter are few. Such are the remarks on the Pythagorean symbolism of the letter Y [And. 1. 1. 27. p. 17), a matter that interested various commentators of the ninth century,^ who found it, e. g., in Isidore Etym. 1. 3. 7 and Servius on Aen. 6. 136, 479. A satisfactory note on "centu- ria" [Phorm. 1. 4. 53, p. 194) may, as Schlee (p. 130) points out, have been suggested by Isidore.' The reckoning of a drachma as the equivalent of three '■'■denariV is not ancient [Eun. 1. 2. 89, p. 169), nor is the notice of Omphale, queen of the Egyptians (Eun. 5. 7. 3, p. 258). This last is based doubtless on Servius (Dan.) on Aen. 8. 300, who in an enumeration of the exploits of Hercules has just mentioned Egypt as the country of Busiris, and says merely, after recounting the theft of Apollo's tripod, "ob quod iratus luppiter eum Omphalae servire praecepit." Possibly the author turned to some other source, possibly he needed no further hints to invent the following: "Omphala regina fuit Aegyptiorum, ad quem dum venisset Hercules, (et) esset ab ea hospitio susceptus, amore illius est captus, promisit tandem ei, quod securum eum permitteret dormire, si vestimenta virilia deponens sua feminea indueret, sicque in contubernio feminarum maneret, quod ita et fecit." Had the author known Donatus, he would have placed Omphala in Lydia, where she belongs. Similar absurdities in the shorter comments are not infrequent. "Sodes" is glossed "socie" {And. 1. 1. 58, p. 23), being appar- ently an abbreviated form of "sodalis,"* "Paba" is only another form of "fabula" {Eun. 2. 3. 90, p. 195); "Dispudet sic mihi 1 For a fuller list see Wessner op. cit., p. 217, n. 1. 2 See e. g. Hagen Anecd. Helv., p. cxxxxiii. In the present case the commentator has altered the usual statement under the influence of the text before him, as the phrase, "quae impedimento aetatis et metu magistri recta est," shows. 'The fact that a most curious comment, displaying the perverted imagination ger- mane to this scholiast appears in D G B while D M Br. give the Isidorian note, indicates that D is conflate here and that the improved note of the second commentary is one of the few later additions in the Halensis. " '■ M. Haut. 5. 1. 14 (Schlee, p. 124) : "voltus quoque hominum fingit^ imitando Br. F. pantomimus poterat dici, pantomimus omnium imitator vultuum est": EM. E is here conflate; see above, p. 377. Ad. Prol. 6: " Synapoth7tescontes Diphili] Sinapothnes comedia est contes difili. Contes difilus grecus comedus fuit, qui compo- suit fabulam quam appellavit sinapothnes, quae latine sonat com- morientes," Br. (II. 4), F (which has " commorientes signat"), C G E M (Schlee, p. 149). "commorientes, nomen properium est. crvvairoOvrjcrKa) (^crwairod- v^ffKo/jLai?) graece est, latine commorior, inde participium. Diphi- lus graecus comoedus fuit, qui scripsit fabulam, quam appellavit synapothnescontes," etc.: D. Schlee is wrong (p. 43) in attributing this curious error of Com. Br. merely to a scribe and not to the scholiast. The earlier scholiast found a wrong division in his textormade one — "synapoth- nes contesdifili" — and manufactured an appropriate comment. The author of the later commentary, who knows at least the form of the Greek participle, is able to correct the blunder. If Schlee's report is exact, M, unlike D, preserves the original comment and omits the later, an indication of its occasionally conflated nature. The later commentator, as we should expect, gives evidence of much wider reading than his predecessor does. While the former appeals to no wider circle than the grammarians, Virgil, Servius, Orosius, Horace's Art of Poetry (?), Porphyrio (?) for help in interpreting Terence, the latter besides depending on those authors, cites also Livy, Cicero, Ovid (??), Sallust, Jerome, Isidore, Horace's Odes.^ It may well be that of the commentators on Terence, Eugraphius, Donatus and the Bembine scholia were used, but the question of the author's sources may not be profitably studied until the exact text of the commentary is ascertained. The possibility of citation at second hand must also be considered. If the reference to Horace's Odes is at first-hand, and an inte- 1 See Schlee, pp. 61, 76-78, 133. Early Mediaeval Commentaries on Terence 385 gral part of the commentary, the date of the work is the latter part of the ninth century, since as Vollmer has shown, the Odes were not known in France until that time.' They came into vogue through the Irish tradition, and that factor is suggested likewise by the commentator's knowledge of a little Greek. His varied quotations prove him a man of humanistic tendencies. We might regard him tentatively as a member of the circle of Eric of Auxerre. Eric was perhaps familiar with the Brunsian scholiast's explanation of "genium = voluptatem,'"' for in his poem on St. Ger- manus IV 510 he glosses "genium curare memento" with: "Quieti et voluptati operam dare. Contra Terentius: suum defraudare genium." The phrase "genium curare" Eric got from Horace Odes 3. 17. 14. We may note also that the gloss " providentia " on " provincia," '' occurs also in the early mediaeval scholia on Juvenal which mention "magister Heiricus."* A MS worth examining is Paris 7900 A s. X. It is well known to editors of Horace's Odes and of "Aero." The scholia on Horace should be compared with those of Terence, which belong, if my conjecture is correct, to Com. Mon.,^ and with the scholia on Lucan, with those on Juvenal, and with the commentary on Martianus Capella which also are found in this MS. Everything points to the entourage of Eric, or his pupil Remigius. Further evidence is necessary before we can decide whether the scholia were inserted in a S or a 7 MS. Dziatzko" agrees with Schlee in placing the commentary on the former branch, but one bit of evidence on which they rely is inconclusive. The order of the plays in M is that of 7, though a note appears in the margin before the Hecyra: "Haec fabula hie interposita. transi hinc ad secundum folium sequentis quatern^i]>. in eo quoque inve^ne]>ris in formionem.'" Dziatzko notes what Schlee neglected to state, ^Philologus, Supplementbd. X. 2 (1905), 288. ^Fhorm. 1. 1. 10; Bruns II. 177. 'Phorm. 1. 2. 22; Bruns II. 180. ■•See Wessner Rh. Mus., loc. cit., p. 208, n. 1. " Qeppert, who partially describes the Paris MSS of Terence (Serapeum XII [1851] 369 ff.) states (p. 372) that in 7900 A vita (III* ) stands after the plays; possibly, then, the scholia have no connection with it. 'Dziatzko Jahrb., loe. cit., p. 433; Schlee, p. 30. 386 E. K. Kand namely, that the commentary on the Phormio begins really on the fourth, not on the second folio of the following quaternion ; the note existed, therefore, in the original of M and was unintelli- gently reproduced by the scribe. In any case, however, it means merely that the author of the note observed in the Commentary a different order of plays from that in some MS with which he was familiar. A theory that Com. Mon. was written for a MS of the mixed class would explain both the order of the plays, and any B elements in the commentary. But this, once more, is a matter for further study. A third commentary on Terence, the Expositio (Schlee's Com. Eec.) has been analyzed in a penetrating fashion by Sabbadini.' He shows the mediaeval character of the preface, its connection with that of Com. Br. (therefore with Vita HI' also), and with the Carolingian scholia on Horace's Art of Poetry. Dziatzko'' had pointed out that it is probably earlier than the eleventh cen- tury, to which Schlee had assigned it, and Sabbadini places it within the Carolingian Renaissance. The vita which begins, "Legitur auctor iste Africanus," is described in certain Paris MSS to Eugraphius, and at the end of the Expositio in Cod. Barber. VIII 47 s. XIII / XIV occur the words: "Quoniam Eugraphius, Donatus, Romgius" ( = Remigius) .' Wessner* asks whether Remigius of Auxerre may not be the author of the commentary. In case he is, the supposition that Com. Mon. proceeds from the circle of Eric receives additional support in view of what we know of the relation of these two scholars.* As sources for the commentary (as distinguished from the vita?) Sabbadini* excludes the Bembine scholia, Schlee's Com. Ant. (i.e.. Com. Br. and Com. Mon.), Eugraphius and Donatus. Sabbadini makes this assertion without going into details, and the matter deserves fresh investigation. The commentary, he shows, furnished material for the lexicon of Papias in 1063, and enjoyed a sovereign authority from the twelfth century to the iStttd. ital. V. 322 ff. ^Stud. Hal. II. 37 N. 2. 2 D. L. 1894, 432. * Rh. Mus., loc. cit., p. 223. 'See Traube's Quellen und Untersuehwn^en zur laf. Philol. des Mittelalters I. 2 (1906), 96 ff. ^Stud. ital., V. 327. Eaely Mediaeval Commentaries on Terence 387 fifteenth, putting even Eugraphius and Donatus into the shade. In the fifteenth century it yielded the field to the latter only after a long and persistent struggle; its career suggests that of another famous mediaeval book, the Doctrinale of Alexander de Villa Dei. In early printed editions the commentary received the distinction of being attributed to Servius. From the MSS cited by Sabba- dini it would not be difficult to construct a satisfactory text ; the list would not include the later folios of D, from which Schlee drew, as Sabbadini shows that the scholia they contain belong to the commentary of Magister Jacobinus of Mantua, who wrote in the thirteenth century. With regard to this and other Terentian commentaries of the later Middle Ages, I may leave the reader to Sabbadini's excellent articles. This paper, I admit, contains many ifs and queries. But I think it right to raise them at this point, leaving others to verify, modify, or refute. Having gone thus far, I am tempted to suggest, partly by way of summary, a few points concerning the introduc- tion of Terence into the Middle Ages, again leaving to others the task of investigating these matters further. The early Carolingian period seems to have had no knowledge of Terence. He is not mentioned in Alcuin's famous description of the library at York,' nor, so far as I can find, in the earliest catalogues of other mediaeval libraries. The Carolingian poets edited by Dttmmler apparently do not mention Terence or show his influence in any striking manner. Manitius thinks that he was known to Aldhelm, Bede, and Alcuin,'' but this opinion is supported by very few citations, which these writers possibly got at second-hand from Priscian or the other grammarians. These works were in the library at York,' together with Virgil,* and Orosius,* — most of the material, that is, on which the Brunsian scholiast depended. Then two MSS of Terence were brought up for Charlemagne, or not long after his reign, from Italy, or it ^Poet. Car. Lat. ed. Dflmmler I. 203. ^Philologvs LII (1893), 547; Wiener Sitzungsberichte CXII (1886), 563 ff. 'Op. ctf.,v88. 1544 ff.: " Artis grammatioae Tel quid scripseremagistri | QuidProbus atque Pocas, Donatus Priscianusve | SerTi