Printed by R. & R. CLARK, LIMITED, Edinburgh.

General Library System

University of Wisconsin-Madison 728 State Street

Madison, WI 53706-1494
U.S.A.

APR 2 3 1906

мем

A YEAR ago I had no design of publishing or composing any such work as this. I knew indeed that the current texts of Juvenal, though praised in reviews and seemingly acceptable to readers, were neither well founded nor well constructed, and that this classic, like many more, had suffered some hurt from the reigning fashion of the hour, the fashion of leaning on one manuscript like Hope on her anchor and trusting to heaven that no harm will come of it. But I neither realised the extent of this injury nor fully understood its causes. I ascribed it firstly to the sloth and distaste for thinking which are the common inheritance of humanity, and secondly to that habit of treading in ruts and trooping in companies which men share with sheep. I did not know that it had also a third source, sheer ignorance of facts, and that the editors had left undone the first of all their duties and neglected to provide the author with an apparatus criticus.

In October 1903, having been asked by Mr Postgate to undertake the recension of Juvenal for his Corpus Poetarum, I began to gather from printed sources the recorded variants; and I soon discovered that Juvenal's modern editors were ignorant or regardless of even the printed sources. I consulted the oldest MSS in the British Museum, but there was little to be learnt from these; so returning to the published records I chose out seven authorities which seemed to emerge above the crowd and to possess some value of their own. Two of these, thanks to Mr Hosius, were collated already; two were in England, so I examined them myself; three were abroad, but of these I procured enough knowledge for my purpose. Their testimony, with that of the Viennese and other fragments, I have added to the witness of our prime authority the Pithoeanus; and I now present to the readers and especially to the editors of Juvenal the first apparatus criticus which they have ever seen.

What stands at the foot of the page in Mr Buecheler's edition

V

is hardly more than a collation of P. It is valuable, but even as a collation it is marred by defects and superfluities. Its defects are trifles, and consist in matters orthographical: we learn from Mr Buecheler when P has aput for apud and haut for haud, but we do not learn when it has set for sed or quod for quot or adque for atque. Its superfluities on the other hand compose the greater part of it; for its greater part is its record of readings imported by the corrector or correctors p. p should be cited, as Jahn cited it, for one purpose and for one alone. Where P is obliterated, there the reading of p affords a means of judging what the reading of P was not everywhere else it is worse than useless. For p is in effect a bad MS of the common class, and it misrepresents that class; and nothing has more obscured the textual criticism of Juvenal than the practice of comparing and contrasting P, not with a good sample of the inferior family, but with one of the worst that can be found.

An apparatus criticus Mr Buecheler's notes are not. He writes on p. xvi

plurimos Iuuenalis codices ipse perlustraui aut ab aliis ita descriptos accepi ut potestas fieret iudicandi, sufficiunt exemplo quorum lectiones Car. Hosius collegit in apparatu critico (Bonnae 1888) et in museo rhen. XLVI p. 287 memorauit, cumque quod frater Pithoeani habuit peculium id Pithoeano paene totum indidisse correctores uiderem, de P et p diligenter rettuli, ceteros omisi nisi ubi horum adiecta memoria declarari posse aut compleri putabam saturarum a Nicaeo ad Heiricum historiam. ceterum nisi nouae chartae emerserint, hunc dilectum codicum ad Persium et Iuuenalem tantidem esse arbitror quasi omnis omnium discrepantia sit adnotata.

To examine 'plurimos Iuuenalis codices' was not necessary: it was enough to cast an understanding eye on the collations of Hosius and especially on the excerpts of Jahn. The chief profit accruing to Juvenal from my apparatus is not derived from 'nouae chartae' (except indeed that the Oxoniensis preserves some thirty verses nowhere else surviving) but from Jahn's Vrbinas 661 and Parisiensis 7900a. How much Mr Buecheler is deceived in thinking that the truth peculiar to the vulgar MSS has 'paene totum' been amassed by p, will be clear to any one who compares my text and notes with his here I will only remark that at III 109 he adopts from p the barefaced interpolation est nec, which not a single decent MS contains, while at x 155 he does not even mention the evidently genuine acti, which is in almost every decent MS excepting P. In one case Mr Buecheler is not himself ignorant of the facts but has kept his readers ignorant.

On p. xiii he says 'Vindobonensem 107 (phil. 128) saeculi noni uel decimi accuratissime conlatum habui ab Alfredo Holdero amico.' It is a pity that Mr Holder should have taken so much trouble; firstly because the MS was fully collated so long ago as 1859, secondly because the only use Mr Buecheler has made of his accurate collation' is to conceal it from the public. The Vindobonensis, between 1 and v 96, not only shares with P fifteen genuine readings which the public believes to exist in P alone, but has half-a-dozen genuine readings which are not in P; III 18 presentius, 67 trechedipna, 168 negabis, IV 3 aegrae, 4 aspernatus, 147 cattis. Mr Buecheler adopts some of these readings and mentions others, but the Vindobonensis he does not mention, and at III 18 he makes believe that praesentius is nothing but a conjecture.

There have been three serious attempts to compile an apparatus criticus. Ruperti in 1801 collected the variants cited by his predecessors and collated eighteen German MSS himself; but few of these are ancient, none seems good, and none is so collated as to be fully known. Next, in 1851, came Otto Jahn, who printed Bertin's minute though not exact collation of the Pithoeanus, and added excerpts from eight other MSS of which two at least are important. These excerpts however are not only meagre but grossly and almost incredibly inaccurate : Jahn's merits were the judgment with which he singled out a few good witnesses from the huge array of Juvenal's extant MSS and the learning and industry he bestowed on collecting citations of the poet from ancient authors. Lastly Mr Carl Hosius published in 1888 an 'apparatus criticus ad Iuuenalem' in which he gave a full collation of six MSS and four florilegia. The title is a misnomer, but the work had great value, since it enabled us for the first time to survey the common text in its entirety and to compare it throughout with P's. But for Mr Hosius' collation of Monacensis 408 and Leidensis 82 I could hardly have set about constructing the present apparatus.

The principal MS of Juvenal is the Pithoeanus, now 125 in P the medical library of Montpellier, written about the end of the 9th century, first employed by P. Pithoeus in 1585, collated by J. V. Bertin for Jahn, and more accurately by F. Buecheler in his edition of 1893. Its original readings are often effaced by the corrections of later hands from the 10th century onward, p

b

but this damage can sometimes be repaired from three fragmentary sources which are closely allied to P. One is the Arou. scidae Arouienses, five leaves of the 10th or 11th century found at Aarau and brought to notice in 1880 by H. Wirz in Hermes XV pp. 437-448. These preserve the verses II 148–155, III 6-13, 35-92, vi 136-193, 252–368, 427-484, VII 57-172, written on pages which tally in size and shape with P's, which contain like them 29 verses apiece, and each of which contains the same 29 verses as the corresponding page of P. A second source is the flor. florilegium Sangallense contained in the codex Sangallensis 870 Sang. of the 9th century and collated in 1885 by C. Stephan in the Rheinisches Museum XL pp. 263–282. Third and most important S are the lemmata, S, of the ancient scholia, Z, preserved in duplicate by this same cod. Sang. 870 and by P itself. These lemmata often disagree with the scholia to which they are prefixed, but they coincide remarkably, though not exactly, with the text of P, and they now and then excel it. Where S or Sang. or Arou. agrees with P, there we have the reading of an earlier MS from which P was directly or with little interval derived.

Over against P and its small cluster of kinsfolk stand the several hundreds of Juvenal's vulgar MSS, dating from the 9th century to the 16th, infected one and all with a plague of interpolation from which P and its fellows are exempt. But about half-way between the two camps lies a considerable fragment, older it appears than P, in which the malady has made Vind. no very formidable progress, codex Vindobonensis 107 or CXI, of

A

the 9th century, containing the verses I 1-11 59 and II 107–v 96, collated by A. Goebel in the Sitzungsber. d. k. k. Akad. d. Wiss. philos.-histor. Cl., Vienna 1859, pp. 37-75.

Then come the seven MSS which I have chosen out of the multitude, AFGLOTU.

Monacensis 408, in the royal library at Munich, of the 11th century, collated by C. Hosius in 1888.

F Parisiensis 8071, Jahn's f, in the national library at Paris, of the 10th century, lacking the verses I 1-III 316 and Ix 40-150, examined for me by Mr Charles Samaran and afterwards in a few places by Mr Louis Brandin.

G Parisiensis 79004, Jahn's g, in the same library, of the same century, similarly examined by Mr Samaran and Mr Brandin.

L Leidensis 82, in the public library at Leyden, of the 11th century, collated by C. Hosius in 1888.