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. BENZO OF ALEXANDRIA AND CATULLUS By William Gardner Hale In an article entitled " Bencius Alexandrinus und der Cod. Vero- nensis des Ausonius," Rhein. Mus. LXIII (1908), pp. 224-34, Sab- badini contributes interesting and important particulars about a figure in the group of little-known North-Italian precursors of human- ism, who, in the early part of the fourteenth century, occupied them- selves with Latin literature. I shall have occasion in this article to differ at several points from Sabbadini. But I should be sorry to be thought not to share in the gratitude and admiration which all Latinists owe him for his con- tributions to the history of humanism. He seems to me, in fact, not only enviably fortunate in having at his immediate command the rich store-houses of Italian libraries, but equally enviable for his gift of the power of drawing their lessons from them. One may well apply to him Horace's verse, Di tibi divitias dederunt artemque fruendi. The first volume of Benzo's Universal Chronicle (preserved to us in the fourteenth-century MS Ambros. B. 24 inf.), which alone remains out of three volumes, indicates, as Sabbadini's examination of it shows, a broad acquaintance with classical as well as mediaeval authors. Among the former, Benzo knew Catullus. I reproduce from Sabbadini's article the passage containing the reference to Catullus, and five of Sabbadini's footnotes. Two of these form the text of my article, while the others are interesting as throwing light upon Benzo's aims. The numbers in the quotation refer to these footnotes. Bencius durchwanderte vor 1315 das ganze Nord-Italien, zB. Pavia und Ravenna, Como, Bologna, Milano, Acqui, Parma, Verona, in deren Archiven und Kirchen er die Inschriften, die Chroniken und allerlei Urkunden bald auszog bald ausschrieb.^ Und nicht nur die mittelalter- lichen Autoren erforschte er, sondern auch die antiken; er suchte zB. Apuleius, von dem er zwei selbst dem Encyclopadiker Vincentius Bellov. [Classical, Philoloot V, January, 1910] 56 Benzo of Alexandria and Catullus 57 unbekannt gebliebene Werke fand;^ suchte, obwohl umsonst, die zweite Dekas des T. Livius* und andere Schriftsteller, die er bei Servius nament- lich aufgefiihrt angetroflfen.* Besonders ergiebig waren seine Porschung- en in Verona, wo er Catullus* und vom Mansionarius Johannes, wie ich glaube, die Historia Augusta erhielt und aus dem hochberuhmten Domarchiv " in quo erant libri innumeri et vetustissimi " (f . 145^) einen Codex des Ausonius hervorzog.* Benzo then obtained a copy of Catullus for himself (so I under- stand "erhielt") in Verona. As for Ausonius, Sabbadini thinks (p. 233) that Benzo probably carried off the MS with him, since there is no reference to Ausonius in the Flores or in the Viri illustres of William of Pastrengo.' The footnotes are as follows: 1. Nach einem Verzeichniss mehrerer Autoren fiigt er hinzu: quorum omnium auctorum libros seu chronicas vel scripsi vel partim seriose partim perfunctorie legi. 2. Huius Apulei duos se repperisse libros dicit Vincencius, unum scilicet de vita et moribus Platonis, alium qui intitulatur de deo Socratis. Ego vero alium eiusdem Apulei librum legi, qui intitulatur sic: Apulei Platonici Floridorum ; alium quoque librum eiusdem comperi qui intitu- latur Asini aurei vel secundum alios intitulatur sic: Lucii Apulei Plato- nici Madaurensis methamorfoseos liber (f . 280). 3. Quod (lies quid) autem Livius hie ponat (iiber Carthagos Grttndung) non legi ex omnibus tribus decadibus, scilicet prima tertia et quarta; puto autem in aliis haberi et maxime in secunda, que nusquam haberi dicitur (f . 135^). 4. Servius . . . . De civitatibus autem tocius orbis multi quidem ex parte scripserunt, ad plenum tamen Phtolomeus grece, Plinius latine; de italicis etiam urbibus Virgilius (lies Hyginus) plenissime scripsit et Cato (ad Aen. VII 678). Actor. Ego vero cum hiis caream auctoribus, sequar Ysidorum, Solinum, Egesippum, losephum, leronimum, Miletum, Librum proprietatum et novissimum omnium Brocardum (f. 126''). 5. Ueber den Larius lacus: "Dicit preterea Catullus poeta Veronensis ad amicum Aurelium scribens sic: Poete tenero meo sodali velim occilio papire dicas veronam veniat novi relinquens domi menia lariumque litus " (f. 94, Catull. XXXV 1-4). Z)omi ist Schreibversehen; ebenso ad Awre- lium % Bemerkenswerth occilio, eine besondere Lesart des Oxoniensis. 1 A similar fate, at a later date, may have befallen the lost Verona MS of OatuUus, and it may hare disappeared from the world while in private possession. We of the present generation know of the taking out of MSS, with completely good intentions, from a library, and of their destruction by fire while in the hands of the borrower. 58 William Gabdneb Hale The short citation in Note 5 (which I assume to contain the only one from Catullus that occurs in the MS), seems to me of great value from two points of view, as will appear below. Occilio is, as Sabbadini says, the reading of O only. Not only R (the Codex Romanus) but, as my apparatus shows, all the other MSS of the existence of which I have been able to get knowledge,' have the reading cecilio (saving a few which present the obvious deri- vatives cerilio, cetilio, cecili§) . Sabbadini puts the composition of the Chronicle between 1313 and 1320, and assigns about 1330 for the date of the author's death. Accordingly, the MS from which Benzo made his extract cannot have been O. It may conceivably, of course, have been a MS lying between and Ver. (I use this abbre- viation of Veronensis for the "lost Verona" MS, to distinguish it from the Vatican MS known as V.) That is, it may have been a copy of Ver., and from it itself may have been copied. Or, it may have been a copy of Ver. having no relation to 0. Or, again, it may have been Ver. itself. I return later to this question. Before drawing the inferences suggested by the citation from Benzo, let me repeat certain convictions which I have several times published : The frequent agreement of G and R against shows that these two MSS cannot have been separate copies of Ver., but must have been copied from one and the same descendant of Ver.; and this descendant, which I shall speak of as alpha, was in all probability a copy which Gaspar de' Broaspini caused to be made from Ver. for Coluccio's use (this copy, not G, being the one that was finished on Oct. 19, 1375), in accordance with the well-known letters.^ Our existing MSS are all descended from OGR, and mainly from the last named. In the quotation by Benzo, domi is, as Sabbadini says, " Schreib- versehen" (for Comi). But for "ad amicum Aurelium scribens" I have a different solution to offer. 1 In an article entitled "The Manuscripts of Catullus," Classical Philology (1908) III, pp. 233-56, I ha-ve published a list of 115 MSS of which I have full collations, and five others belonging to the sixteenth, seventeenth, and eighteenth centuries, upon which I have memoranda. These collations were made, with the invaluable help of my research assistant, Mr. (now Professor) B. L. UUman, in pursuance of a plan which I announced in an article entitled "Catullus Once More," published in the Classical Review for April, 1906. 2 Novati, Epistolario di Coluccio Salutati, I. 207, 222. Cf . 1. 170. Benzo op Alexandria and Catullus 59 has interstices before a number of the poems, and against the first verse of a number of others it has two light lines, generally overlooked by editors,' in the margin. Before XXXI (and here only, in its whole compass), it has, in addition to the two light strokes in the margin, a formal paragraph mark. With few excep- tions, G corresponds, but has a formal paragraph mark where O has light lines. R has an interstice wherever O has either an inter- stice or the lines. It is clear that the arrangement in a and in Ver. was as is vouched for by the substantial agreement of and G, and that the scribe of R, on the other hand, was working under special intentions or instructions. In the article in Classical Philology referred to, I have shown that there were but a few titles in Ver., and that most of the titles which appear in G and R were devised by the "corrector" of R (R^), namely its owner, Coluccio.^ Before our poem (XXXV), as elsewhere, has no title. R^ wrote in black ink, in the margin, the temporary title Ad libellum suum de cecilio (which comes down in the GLA gioup, with the spelling cicilio in LA) , and, in the interstice, the same hand wrote, in red ink, the revised title Ad Cecilium iubet libello loqui. G" wrote, in red ink, the title Ad Cecilium iubet libello loqui in the margin. The general evidence makes it most improbable that either of these titles was in O, or even in a. But, in any case, neither of them could have suggested ad amicum Aurelium scribens. A different explanation must be found. The one which at once suggested itself to me appears convincing. Let us turn back in until we reach an interstice. This appears six and a half pages earlier, before poem XXI. The same interstice appears in G, as well as in R. There is no other interstice between XXI and XXXV in G. It follows from these facts that there was an interstice before XXI in Ver., and that this was the first interstice back of XXXV. Now the first verse following our interstice is Aureli pater exuricionum {-tionum R). Here is where Benzo got his "friend Aurelius." The marginal marks of division in Ver. had not saved him. They were very likely unobtrusive, as they are in O. Benzo may not 1 Merrill (Catullus, 1893) notes nearly all. ^ As I showed in the same article, G got its paragraph marks, like most of its titles, Tariants, and corrections, from M, one of the daughters of R. But the argument is the same. 60 William Gasdneb Hale have noticed them at all; or, if he noticed them, he probably thought them simply indications of subdivisions in a long poem. Seven signs with precisely this intent appear in LXII in GR and M, formed in R and M of light strokes meeting at a right angle at one end, and taking in G the form of regular and clear paragraph marks. In passing, I may say that Benzo's citation practically proves that there were no titles in Ver. between XXI and XXXV, a con- clusion which harmonizes with the view which I had independently reached for most of the poems, including XXI-XXXV. The poem from which Benzo inferred the friendship of the two men is an abusive one. With a break of only one poem, it is pre- ceded by two others, XV and XVI, of a similar character. Catullus made his attitude toward Aurelius plain; but Benzo does not know it. There are many changes of subject, and many different persons, gods, or places addressed, between XXI and XXXV — ^Varus, Furius, luuentius, Thallus, Furius again, the poet's cupbearer, Veranius and FabuUus, Romulus, Alfenus, Sirmio, Ipsicilla (?), Vibennius and his son, and Diana. Benzo did not know this; for, if he had, he could not have supposed that this long succession of pages formed a single poem, addressed to Aurelius. On these facts I base an important conclusion, which, while not inevitable, is highly probable, namely: Benzo did not really know Catullus. In the only part where we have any indication that he had any acquaintance with his author at all,' he had read only the poem from which he cites. He had not read XV or XVI. From XXI to XXXIV he had not read. He had looked at his author, not "seriose," but "perfunctorie." But (and here is my second conclusion) a man who knew his author so little can hardly have possessed a manuscript of that author. It is much more likely that Benzo was turning over "perfunctorie" a manuscript in the possession of another person, or of a library. We know pretty surely that there was a MS of Catullus in the Cathe- dral Library in Verona, and that Benzo rummaged about in this library (he himself says of it, "in quo erant libri innumeri et vetus- 1 For his method, compare his own account, in Sabbadini's note 1, of his treatment of the historians and chroniclers from whom he drew: "vel scripsi vel partim seriose partim perfunctorie legi." Benzo op Alexandria and Catullus 61 tissimi"). It is altogether probable, then, that Benzo used the lost Verona MS itself, in the Cathedral Library. This brings us to a summing-up of the meaning of Benzo's citation from my first point of view. It appears to be commonly assumed, as by Sabbadini in the case of Benzo, that a man who, in the four- teenth century, cites an author (unless the same citation appears in a Latin classic) possessed a copy of that author; and there appears to be a disposition to reconstitute private libraries on this basis. I have always questioned this position on general grounds; but, so far as I know, Benzo's citation affords us the first means of testing it. It does not bear the test, and the position must be abandoned. An application may be made in a particular case. On p. 1 of Le Scoperte dei Codici Latini e Greet ne' Secoli XIV e XV, Sabbadini says of the poem Ad patriam venio that it "dev' essere stato scritto dal Campesano suUa sua copia di Catullo." I doubt that Benvenuto possessed a copy. I am more inclined to think that his verses were written at the end of the Verona MS, either by Benvenuto himself, or, less probably, by some one who had a copy of his verses. They are in both G and R (in the latter, at the beginning), and were, therefore, probably in a. It is more likely, on general grounds, that the copyist of a got them from the MS which lie was copying than that he hap- pened, just at this moment, to have another copy of Catullus at hand.' In addition, I have shown (" Classical Review," loc. cit.) that the note at the end of G was copied (compare Chfttelain's previous suggestion) from a ; and in this note the scribe (of a) explicitly says that he had no MS of Catullus at hand except the one from which he was copying. There is nothing improbable in the hypothesis that an early four- teenth-century scholar felt free to enter extraneous matter in a ' Sabbadini, op. cit. 4, says that the Veronese "Gaspare de Broaspini .... aveva un Catullo." The evidence relied upon (the letters of Coluccio) does not seem to me to justify the conclusion. All that appears from the letters is that there was a MS within reach of Caspar, from which Coluccio wanted a copy made. This may perfectly well have been the MS in the Verona Cathedral Library. But the fact that Coluccio had a copy made from the MS sent him, instead of simply keeping that MS, suggests another possibility, — which I put forward as no more than such. Caspar, having a copy (our a) written (quite conceivably correcting it himself, as Coluccio corrected his own MS), and knowing that Coluccio was willing merely to borrow for the purpose of transcribing and had abundant scribes at com- mand, may have asked Coluccio to return the copy to him. If this took place, and the copy reached him in safety, he then "possessed Catullus." 62 William Gardner Hale library MS which he was reading; for there is proof that marks of some kind, at any rate, were made in this particular MS by such a reader or readers. Against five of the seven passages included by Jeremy of Montagnone in his Compendium Moralium Notabilium (all except LXVI, 15-16 and LXVIII, 137), R has in the margin either a pointing hand or a brace or a No {=nota bene). This cannot be mere coincidence. The only reasonable explanation is that these indications of special interest came down through o from Ver., that is, that the scribe or corrector of a repeated, in one form or another, the indications which he found in Ver., and that Coluccio (for the work is his) again repeated them from a when he corrected his MS R. As for the origin of these marks in the margin of Ver., the probability is that Jeremy himself made them in preparation for his book of selections. Of course it is possible that somebody had made the marks before Jeremy read the MS; though it is less likely that he would accept the choice of someone else than that he would make his own. But, in any case, these marks, no matter who made them, clearly served him for a guide in taking his excerpts. This view falls in with the fact that the passage quoted in the Flores (which is dated 1329) is one of Jeremy's quotations, and one of the five passages out of Jeremy's seven that are marked in R. Probably the author of the Flores, looking at the Verona Catullus for something to quote (and this was after Jeremy's death in 1320-21) had his attention attracted by marks of interest which he found there, and made his selection from the passages thus singled out. It is an interesting fact, too, that the passage which he took is the earliest of Jeremy's passages. There is only one conceivable escape from this conclusion, namely in the h3^othesis that Jeremy himself possessed a copy of the Verona MS and entered these marks upon the copy, and that it is from Jeremy's MS, and not from Ver., that G and R are descended. But this will not serve. It is extremely probable, in the light of Coluccio's three letters, that the MS of Catullus of which he was asking Caspar for a copy was the one which, as we pretty surely know, was in the Verona Library. This is enough to account for GR. To suppose that a MS of Catullus belonging to Jeremy had also come into the possession of this same library is to assume a highly improbable coincidence where a simple explanation is already at hand. Benzo op Alexandria and Catullus 63 The general scepticism which I have expressed above with regard to drawing inferences of ownership is thus confirmed, and made to cover even more ground. Here is Jeremy, with seven excerpts from Catullus! Relatively, this is very strong evidence in favor of possession. Yet we have seen that, in all human probability, Jeremy did not possess a copy, but used the Verona MS. What has thus been established, with as much approach to cer- tainty as is likely often to be attainable in such matters, in the case of two different early fourteenth-century men who cite an author, one of them even in seven passages, should make us weigh evidence very carefully in considering any question of the possession of MSS in the period dealt with. My own feeling is that, in the lack of positive proof of some kind, only the evidence afforded by a real acquaintance with an author on the part of a given man should lead us to credit him with the ownership of a manuscript of that author. This position is in turn somewhat reinforced by a consideration of another kind. All our existing MSS of Catullus, as it seems to me, can be carried back to an origin in OG or R, or (in a few cases) a combination of two of these MSS. There is no trace of the influence of any MS outside of these three. If we had to believe, on the evidence of a citation or an imitation, that not only Petrarch (the case is somewhat different with him, and is complicated) but Benzo, Benvenuto, Jeremy, Mussato, the author of the Flores, and William of Pastrengo (to say nothing of Caspar), possessed each a copy, it would be hard to think that these copies had all passed away without leaving a descendant or a trace of influence among the hundred and fifteen MSS which we still possess. Of course manuscripts disappear, and one who looks for a steady succession of mother, daughter, granddaughter, etc., has to put up with disappointments. But for the three great manuscripts there is no lack of descendants, or evidence of influence, at some stage of removal. It would be a strange disparity if precisely the opposite had happened in the case of every one of seven other manuscripts which would rank with OCR. It should be remembered, too, what difficulty Coluccio had in procuring a copy, either to own or to get transcribed.' 1 It will be seen that I do not incline to Ellis's view that the variants in our earlier MSS are to be assigned to a comparison of early copies of Ver., in the text of which various readings had arisen in the process of copying. 64 William Gabdnee Hale Thus far my discussion has had to do with the question of the dissemination of MSS of the Latin classics in the first half of the four- teenth century. But Benzo's citation has another bearing also. In endeavoring to reconstruct the text of the lost Verona arche- type from 0, G and R, we have, roughly speaking, two traditions to deal with, that of on the one hand, and that of G and R together, as representing a, on the other hand. Other things being equal, we should balance the one tradition against the other. But the question is, Are the other things equal? What is the general temper of the scribe of 0, and what the tempers of the scribes of a, G and R? Did these men aim simply to copy what was before them, or did they try to emend where they thought they found error? Were they careful or were they careless? Were they painstaking in endeavoring to preserve everything, or were they easily content with a part? Our answers with regard to these questions (and they are very difficult ones) will not only be important for the text of Catullus, but will contribute toward an understanding of the general methods of late fourteenth-century scribes and correctors, and so have bearings for the text of other authors. I shall sometime discuss the evidence in full. In the present paper, I am concerned only with what may be learned from Benzo's citation. The agreement of Benzo with O in reading occilio where a, (as proved by G and R) has cecilio shows that Ver. had occilio, and that the scribe of a, or a possible corrector of a, (who, as already suggested, may have been Gaspar himself) , made an easy emendation based upon the occurrence of cecilia (for cecilio) in the last verse of the poem. There is very little likelihood that the scribe of Benzo's extant volume, if it was written after 0, happened to know this particular MS or a now lost descendant of it, and made Benzo's reading con- form. then is shown, at this point, to reflect the lost archetype more faithfully than GR. But another difference of reading is also covered by Benzo's citation, namely in the word menia, which undoubtedly represents what Catullus wrote. has uenia. G had meniam, of which the last letter (the z-shaped m) has been erased. R reads menia now; but above the a a delicate erasure can be made out, which, though no color is left, corresponds perfectly in shape to a virgula in the Benzo of Alexandria and Catullus 65 style of the scribe. OG and R thus agree upon a final m, which Benzo, or the writer of the extant MS of his work, rejected. But did ueniam or meniam stand in the lost archetype? The latter is not an existing word, the former is. A correction from a certain error to an existing and familiar word, even though we now recognize that this is not the right word, is more likely than a correction from an existing word to a non-existing one. On general principles, then, we should say that the archetype read meniam. But we no longer have to rely on general probability. The agreement of Benzo with GR against proves, in connection with the agreement of aijd GR about the last letter, that the archetype read meniam. Then it was the scribe of that was at fault, either through fully intentional emendation, or (with a half glance at the word) half-unconscious emendation, or by simple blundering. In any case, it is here O that misrepresents the reading of Ver., while a preserves it, in spite of its being partly wrong. Thus Benzo's citation, as a whole, proves both traditions to have erred, one in one place and one in the other, and, so far as its sole evidence goes, leaves the balance even.