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. ON A VENETIAN CODEX OF PLINY'S LETTERS By Elmer Trubsdell Merrill In St. Mark's Library at Venice is a manuscript volume of miscellaneous content written on paper by various hands of the fifteenth century and officially designated as Lat. class. XI. cod. XXXVII. An inscription of presentation on the last original page of the book (the binding is comparatively recent) professes to date from 1467,^ but the parts of the book may well have been written and put together at a considerably earlier period in the fifteenth century. The contents of the book are as follows: foil. 1-51, Orationes M. Tvllii Ciceronis; foil. 52-62, Quoddam principium Antonii Lusci super orationes tullii; foil. 63-67, Thimeus tullii; foil. 67-69, eiusdem de fata; foil. 72-91, C. Plinii Secundi Veronensis oratoris clarissimi nonnvMae epistolae Ad Secundum et alios feliciter incipiunt [Plin. Ep. I-II. 17 . 17 aestate nam ante, where the writing breaks off at the end of page and fascicle, but with the catch- word meridiem in the lower margin, showing that once the text was, or was meant to be, farther continued]; foil. 92-93, C. Plinius Romano suo saltdem. Post aliquot annos . . . centum milia contulerim [PUn. Ep. II. 1-4.2, without title of work or book. The writing ends at the bottom of a recto, and the uerso is left blank] ; foil. 94-102, Breuis anotatio de dvobus pliniis tieronensibus oratorilms ex muUis hinc collecta per lohanem Mans. Vef., followed by a columnar index of addresses in Plin. Ep. I (Ad septidum. Ad arrinium, etc.), flanked by a four-line metrical epitaph on either PUny, and this succeeded by the text of the first book, without inscription for either work or book. At the end of letter 22 (letters 8, 12, 23, and 24 are omitted in both index and context) follows the colophon c. plinii secudi epistolarum liber primus explicit. Incipit liber secudus. Then follows Incipiunt capta libri seeUdi, with a columnar index of addresses of letters in the second book, but no text; foU. 106-121, Cronicae martirorum pars. 'See note in Dora Johnson's article, "The Manuscripts of Pliny's Letters," in Classical Philology, VII (January, 1912), 71. [CLASaiCAL PHILOI.OQT XII, July, 1917] 259 260 EiiMEB Tbuesdell Mebkill The intermediate folios for which no contents are noted are left blank. Four blank leaves also precede the first written folio, and fourteen blank leaves follow the last. I first examined the book and made excerpts from it in 1899, and a few years ago my pupil, the late Miss Dora Johnson, at my suggestion collated the Pliny parts for me. With regard to the letters of Pliny on foil. 72-91, I need say merely that the text is of a poor and contaminated type common enough in the fifteenth century and of no critical value. It may therefore be disregarded. But the text on foil. 92-102 (in a different — very crabbed — hand from that of the preceding text and on different paper) stands on quite another footing and deserves attention. With this alone the present article is concerned, and this extract is herein designated as m, the mark that I have assigned to it also in my critical edition of books I-IX of the Letters, the printing of which by Teubner was suspended midway in its course on the outbreak of the present war. I am inclined to think that m is, at least for Book I, an inde- pendent, uncontaminated, and moderately accurate copy of that Veronese archetype of the eight-book family of the Letters which Guarino carried off from "Verona to Venice in 1419. This opinion of mine was recorded by Miss Johnson in her preliminary classi- fication of the MSS of Pliny's Letters referred to above (note on p. 259). That m is an immediate copy (that is, made from the Verona MS itself, and not from a copy of it), I should hardly ven- ture to affirm; but at most only one copy probably intervened between m and the Veronensis. Of the Veronese tradition of the Letters I have treated in an earlier article in this journal,^ and I have indicated there and else- where'' that there is no other source for the extant eight-book tradi- tion of the Letters than this one MS, which had been lying in the chapter library at Verona since at least the tenth century.' But 1 "On the Eight-Book Tradition of Pliny's Letters in Verona," in Classical Phil- ology, V (April, 1910), 175-88. 2 E.g., in my article on "The Tradition of Pliny's Letters," in Classical Philology, X (January, 1915), 8-25. > It may not be out of place to remark here that this conclusion was frankly out of accord with the published impressions of Professor Remigio Sabbadini, to whose On a Venetian Codex of Flint's "Lettebs" 261 Guarino found that it needed much critical revision, and to this he set himself with great zeal. He put off his eager friends, who wanted a glimpse of the new wonder that had brought to light more than twice as many letters of Pliny as had heretofore been known, and when he finally did lend his MS around, I think there is good reason to judge from his own correspondence, and from the condi- tion of the now extant MSS of the eight-book class, that it was his corrected copy and not the original that was supplied. The eight- book text that was spread abroad during the rest of the fifteenth century had generally as its basis the Guarinian recension of the Verona archetype and not that archetype itself. But a few copies of the Verona MS appear to have been made that show no trace of the revising hand of Guarino. One of these gives us m, which is unfortunately but a fragment; a second is D (Dresdensis D 166), which is complete; a third is, according to Miss Johnson's belief, Holkhamensis 396, about which I am not yet ready to express a judgment. It is of course possible that some or all of these, and perhaps other copies not now traceable, were made before Guarino acquired the MS; but antecedent probability is rather against this, and (though I am not a competent palaeographer) I should not be readily inclined, in the lack of expert opinion, to believe that so fine a piece of calligraphy as D, in the restored style of the fifteenth century, could have been executed before 1419, or was hkely to be until the Veronese MS had assumed a position of importance. That m gives the tradition of the text in general agreement with D, but is from a common source with D instead of being a copy of D, is readily to be deduced from an inspection of their comparative readings. A full statement of these is manifestly too long to be included here; but a classified selection will probably be regarded as sufficient to establish the point, if I may be trusted in the afiirma- tion that no evidence looking in any other direction can be detected in the complete luiria lectio. In the first part of the ensuing state- ment I intend to confine myself to the text of Book I only, reserving attention I called it. I am now interested to note that in the new edition of his Storia e Critica (Catania, 1914) he has revised his judgment, and is assured that the letter of Guarino announcing his acquisition of the MS was written from Verona in April-May, 1419, and not from Venice. 262 Elmer Tbuesdell Mbrbill to a later position in this article some consideration of the readings of II. 1-4.2, the text of which (on foil. 92, 93) immediately precedes in the Venetian codex the Breuis Adnotatio de dudbus Pliniis and the text of Book I, with the indices. As a preliminary it should be noted that m omits in both index and context I. 8, 12, 23, and 24, being herein in agreement with D, but differing from many other MSS of the eight-book class which in this as in other respects display manifest signs of interpola- tion from the F-tradition. My first purpose is, then, to show that D and m are in striking agreement in details of text as against all other MSS noted in my apparatus criticus (BFMVIiLH, etc.), except in some degree those (oux) of the eight-book family which are more or less contaminated from the F-tradition. And in order to mark the distinction between the MSS of pure eight-book descent (Dm) and the interpolated MSS of the same family (oux), I have limited the statement of identical readings to those in which Dm stand entirely apart even from these interpolated relatives of theirs, thus excluding from the lists even the very numerous instances in which the reading of Dm differs, indeed, from readings of MSS of other families, but is con- curred in by one or more of the group oux. (In this excluded class of readings the agreement of one or more of oux with Dm is of course to be explained as due to the retention by oux in the given cases of the genuine reading of the Veronese eight-book archetype in preference to the available interpolation from the F-tradition.) In order further to strengthen the proof of the close and ex- clusive relationship of D and m, the readings to be cited are con- fined to those where the agreement of the two MSS is upon manifest errors. It will be remembered that the examples are all taken from Book I of the Letters. Citations are by page and line from Keil's critical edition of 1870. In the following instances D and m agree as against all other MSS (even oux) in wrongly omitting from the text the word or words indicated (I have already remarked that they agree in omit- ting altogether I. 8, 12, 23, and 24) : 4.8 nostri 5.6 tamen 4.19 sed sane blandiantur 5.8 possident-9 partiuntur On a Venetian Codex of Pliny's "Lettebs' 263 5.10 possident-12 enim 21.11 uale 5.14 hoc otium (om. et Flor. 22.8 ut {post ipsius) Mor. Auct.) 23.1 nostro 8.12 nostri 23.8 L. 9.2 expecto- prudens 23.9 et 9.25 Ioui-26 aviveufffv 24.11 et {altero loco) 10.4 nostrae 26.4 non 14.20 magis miror quia magis 26.23 mihi 15.16 an- 17 licet 27.14 iUa 15.21 me 27.21 et magnifica 18.17 iubent 28.2 quia 18.27 a 29.25 eius 19.24 Macrinus 30.5 ipsam 20.1 enim 30.17 est 20.26 uale In the following list of readings D and m agree in error as against all other MSS (even oux) : 3.7 curatius 8.11 potuisse] puto esse 4.10 acrius enim non tristius 8.15 Modesti] domestici 4.17 misimus 8.18 credulitatem 4.20 commendantur 9.4 constauit 5.3 tuum 9.6 dicta factaque 5.4 platanum opacissimum 10.9 in praesentiam 5.17 sortiuntur 10.12 cui] cum 5.19 enitere] conice 13.20 ille ad inuocationem 5.23 cursulano 13.21 rogauit ita et quo 6.5 deueneris 14.5 musion 6.15 tectior coepit (cepit m) 14.23 tamen mihi datur cernere 6.18 appellata dicit multa 7.1 a domitiano erat relegatus 15.7 honor 7.7 soleant 15.9 par] parum '7.8 iam quidem de 15.10 attentius 7.10 quidem (quid m) est fas 15.22 esse et hanc philosophiae 7.15 fabium rusticum iustum quidem rogat 16.7 at] aut 7.16 subtiliter 18.23 recitationi timet] sim et 18.24 quisque] quis 7.18 ferre diutius sollicitudinem 19.1 non tardat 7.20 coimus] quo imus 19.14 laudibus laudandus affecerit 7.23 regulo nuntiandum (-c- m) 19.18 qui] et 8.1 nundum 19.21 nam ita a me institui 8.6 inhaereret (-he- m) 20.5 patauis 264 Elmer Truesdell Merrill 20.9 industriam 24.23 est ualde uale 20.10 ac iam] acia 25.1 permittit 20.16 imaginosum os 25.3 infingenda 20.20 hie] his 25.5 ietu] situ 20.24 inuenio 25.8 sane] in se 21.3 fericulo 25.12 meliorem est quisque non minus mile 25.14 denique pictorum multorum 21.4 comoedes (-med- m) animalium 21.6 caditanos 25.15 multitudo 21.7 certe an mihi 25.20 rationibus 21.8 nos lus.] uoluissemus 26.5 exemplum 21.9 cautius 26.7 in Verrem] enumeraret 21.11 excuses 26.14 nisi ut in 21.17 quasdam causas 26.16 mouit nee] non 26.17 indicia hominum 21.24 in historia tibi magis 27.2 panta denique 22.3 lepores dulcedines amari- 27.16 delicate tudines amores 27.17 is demum] idem 22.4 mollibus leuibusque 27.18 quae aeque (e- m) de 22.15 requireremus 28.8 illud 22.17 at] ad auespoene 22.21 Iceiano 28.12 illam plenam 22.22 adhoc 28.19 auctoritati tuae credere 23.3 est experiri (experr- D) debeam 23.6 potest] post 28.22 confirma idem] item 28.23 erraro 23.15 paucos 29.4 ne] non 23.19 tu] tuam 29.5 est] et 24.5 id] at 29.11 iam] tam rescribere 30.1 aliorum] alienum 24.13 aduersitas 30.5 sapientia 24.17 equitis 30.14 nam iam impetu These two incomplete lists of exclusive errors by lacuna or by distortion in Dm within Book I alone are amply sufficient to estab- lish the close kinship between them as against all other MSS. If the lists had been made complete, or if there had been included an enumeration of the errors that are concurred in by one or more of DUX, and if a list were also subjoined of the few right readings that are found only in Dm, the presentation would be even more striking (e.g., D and m alone have in 22.3 re uera quales Catullus aut Caluus, which is indubitably right, though the nine-book tradition fails us here). 4.11 me 6.4 me 6.10 magis 7.10 ne On a Venetian Codex of Pliny's "Letters" 265 D and m therefore are closely related in a common tradition. Either one is a copy of the other (or at least in direct descent from it), or they are both from a common source. But D cannot be descended from m, for (1) the make-up of the volume in which m is contained discountenances such a hypothesis (m probably never had more of the Letters than it now contains), and (2) D on careful investigation shows no signs of contemporary conjectural emenda- tion, nor of interpolation from any of the MSS extant at the period of its production,^ while yet there can be cited a number of read- ings in Book I of D that are simply missing in m. For example: 8.1 a me 8 . 26 potest- 9 . 1 infida est 9.11 ipse [arde non) 27.6 latius There are also a number of cases where SVO or S. is omitted in m in the addresses of letters, but these might possibly have been corrected in copying even by unconscious habit, and they are therefore disregarded here. The foregoing list of other lacunae in m properly filled out in D is, to be sure, short (though 8. 26-9. 1 is very striking), for m is a tolerably accurate copy of its original; but in the lack of any indication in D of any recent tendency to conjectural emenda- tion or to interpolation from other extant sources, it is good evi- dence that D was not copied from m. Under this postulate of no contemporary emendation by con- jecture or interpolation, corroborative evidence that D was not copied from m may be found in the consideration of some readings where m shows the wrong and D the right text, and happy chance can hardly be assumed to have been responsible for the result in all or most of the cases. The reverse cases of right readings in m and wrong in D must of course be disregarded, since such errors in D might as well have originated in copying from m as from any other MS. Readings in point are : 6.22 molestus D modestus m 8.8 cui D cum m 7.3 sentias D sententias m 8.9 saeculi D secundi m 8.7 iudicio D offo m (officio r) 9.13 proximo D proximum m * For the later inserted F-readings do not obacure the original text, and may be disregarded. 266 Elmer Truesdell Merrill 9.16 agitatione D cogitatione m 20.15 eius D etiam m 10.9 quid uir {corr. ex quidum) 21.10 aliis D alii m D qui uir m 21.20 mire D michi m 14.2 rumoribus D timoribus m 27.22 tonat D sonat m 19.22 ex D et m 28.24 necessitatem D necessi- 19.23 multum adhuc D adhuc tate m multum m 30.24 laetioribus D latioribus m Therefore, D cannot have been derived from m. The same manner of proof may be adduced to show that m cannot have been derived from D. Following are readings in m that are simply missing in D: 6.6 nos 22.19 uale 9.21 uale 23 . 27 et siquid carius patria 15.4 hoc 26.15 quod 19.3 his 27.10-11 Graeca om. in lac. m. 19.5 ne om. sine lac. D 22.11 cum (ante scripsi) Some of the pertinent instances where D has the wrong reading and m the right (see the remarks above on the converse case) are as follows: nostra liberalibus m nostra non liberalibus D auditor m aditor D reposcatur m reponatiu* D illi m sibi D septitio m septio D et m uel D prauum m paruum D fallacia m f allatiaque D It is to be noted that if the scribes of D and m had had any tendency to emendation, or to interpolation from the well-known F-tradition, the processes would surely have gone far enough to show patent evidences of their action, and the lists given above would accordingly have had a very different appearance. D and m, then, are entirely independent the one of the other, and owe their close likeness to the fact that each is a copy from the Veronese archetype of all the eight-book MSS. The text of m 4.11 uolebamus m uolebam D 14.13 4.14 nostro m non D 5.15 excude m exclude D 19.6 5.16 gestatio m uectatio D 19.8 6.23 regulus m regulis D 20.11 8.8 satrio m satiro D 20.28 satrius m satirus D 21.25 3.23 quot m quot corr. m? ex 22.17 quod D 27.7 On a Venetian Codex of Pliny's "Letters" 267 accordingly not only aids in the reconstruction of the lost Veronese text in the first book, but tends to estabhsh confidence in the integ- rity, even if not in the impeccability, of D in the following books, which are not included in m. For the overwhelming number of extant eight-book MSS are, like oux, or even more so, the product of the Veronese archetype greatly modified in the fifteenth century both by conjectural emendation and by direct substitution from the F-text. But as the resulting product of this scholastic process is, at least in its better forms, more nearly "correct" in the total than the lost Verona MS could have been, a student might, with some degree of reason, have regarded the interpolated tradition as the authentic one (however lacking in practical value at the present day for the reconstitution of PUny's autograph), and D as a blunder- ing variation of it (cf. the history of Notitia and Curiosum). This error is no longer possible; but it was one to which my own judgment was inclined to succumb in the earlier period of my study of the eight-book text. Keil, to be sure, had rightly distinguished the three families of MSS of the Letters, and had designated D as the type of its class; but Keil had not examined many other eight- book MSS, and was in the main content with collations of the early printed editions (prr), now known to exhibit a badly contaminated text. That he pitched upon D in addition seems to be due to a happy accident rather than to a mastery of the complex problem. D was a beautiful and striking codex, and it lay almost under his hand in Dresden. Other examples of the class were less accessible. The problem, indeed, could not have been thoroughly grasped until the researches of Sabbadini into the activities of Guarino and his fellow-humanists were published, and this was later than Keil's day, and than the eariier years of my own studies in this field. It remains to discuss the text of Plin. E-p. II. 1-4. 2, which is contained on foil. 92, 93 of the Venetian codex, preceding the text of Book I (foil. 94-102) that I have treated above. Miss Johnson, as I have remarked elsewhere, was of the opinion in 1911 that this fragment was an integral continuation of the copy that follows it in position, and that the two leaves had been accidentally misplaced in the binding of the book. To me, on the other hand, it appeared in 1899 that the texts of these two parts were too dissimilar in the 268 Elmer Tbxjesdell Merbill marks of tradition to permit the judgment that they could have been copied from the same MS. I am still of that opinion, but I had no opportunity to discuss the matter with Miss Johnson. The determination of the case is of course embarrassed by the brevity of the fragment of Book II, but yet this is long enough to render judgment reasonably certain. I much regret that in my notes made in 1899 I can find none on the resemblance of the hands in the two parts, and none on the make-up of the fascicles of paper in the bound volume; nor do I find any remark on these questions in Miss Johnson's notebook on m, which is now in my possession. When, in 1909, I went to Venice in the hope of repairing some of my sins of omission, I was so unfortunate as to arrive on the very day when the Library was closed for its annual vacation of a fort- night, and I could not remain in the country to await the reopening. The outbreak of war interfered with a projected later visit. At the time of the acquisition of the Veronese archetype by Guarino (1419), no other tradition of the text of Pliny's Letters than that of F was known in Italy, nor did any other become known there till much later in the century. Therefore the text of m in the portion now under discussion may be expected to agree sub- stantially with that of the Verona MS (as exhibited in D) or with that of F — that is, to be at basis either an eight-book or a ten-book text; but it might be, like so many of the Pliny MSS of its period, an eight-book text contaminated (1) by error, (2) by conjectural emendation, (3) by interpolated readings from the F-tradition. Inspection shows immediately that it is at basis an eight-book text; it agrees substantially with D and not with F — witness such instances as the following, where it concurs with D in errors as against F : 33.2 ipsum om. 33.23 nuntiasset (-ci- m) 33 . 7 repararet 33 . 28 consulares (om. -que) colt (corr. w'' w coiit) 34.19 sit {fortasse sic m; corr. 33 . 23 accurrit m^ in si) But it shows in its original text a number of F-readings, where these are different from those of D, and are manifestly wrong. For example : On a Venetian Codex of Flint's "Letters" 269 34.4 fieri Fm flere D 35.8 crebri om. Fm, add. D 34.8 una om. Fm, add. D 35.22 at ow. Fm, add. D 34.10 recentibus om. Fm, odd. D 36.16 aliqua etiam Fm alii 35 . 3 iam igitur Fm amicitur D quam D Accordingly m displays in II. 1-4. 2 an eight-book text inter- polated from F. But this is not all, for an apparently later hand has changed still other readings to make them into accord with F. These, however, I need not cite (a number of such alterations were similarly disregarded in my discussion of the text of Book I). More- over, m farther departs from the following of the eight-book text in admitting readings that are due to such errors or conjectures as were perpetuated in the two earliest printed editions of the inter- polated eight-book type, p (1471) and r (1474). For example: 34 . 10 uariis mp uanis FD 35. 15 uero mp enim FD 35.22 ut audias ueni m ut audias hue ueni r ut hunc audias ueni FD 36.3 simpsedFD 36 . 9 ferunt mr fertur FD Therefore, m in this part must have been copied from some MS of the contaminated type that became common after Guarino and his compeers had worked their well-intentioned but lamentable will on the text of Pliny's Letters. The text of Book I in m shows no such signs of recent contamination. The texts of the two parts could not have been copied from the same MS. I accordingly hold that my judgment in 1899 was correct; that I am right in including in my forthcoming apparatus criticus the readings of Book I in m, and in excluding those of the fragment of Book II; and that Marc. Lat. class. XI. cod. XXXVII ought to be credited in the catalogue of St. Mark's Library with containing three distinct parts of Pliny's Letters instead of two. The bookbinder did not blunder. A few remarks may be appended concerning the text of the Breuis Adnotatio de duobus Pliniis contained in this codex. Since I printed (Classical Philology, V, op. cit.) this interesting screed by lohannes de Matociis, a number of other MSS of it have come within my knowledge, viz.: Marcianus lat. XI. 37 (m) and X. 31; Vrbinas lat. 1163; Vaticanixs lat. 1952 and 1955; Escorialensis N. 270 Elmer Truesdell Merrill III. 10; Mutinensis T. 6. IB, Q. 8. IS, and Q. 5. 23; Sangallensis 896 (?). Doubtless numerous further additions can be made to the list. Of all the MSS at my present command m seems to afford the most sincere text; and on the reasonable possibility that it is, in the Adnotatio, an immediate copy of the author's autograph, I might make on its (not unsupported) authority one or two small changes in my text, such as hinc for hie in the title, and inuenitur for inuenerit in the context (p. 188, 1. 12). A point of interest to me is that m reads natura for materia of other MSS and printed texts (p. 188, 1. 4), natura being an emendation that I had suggested in this place from conjecture (na> ma = materia). University of Chicago