(b) Haut. IOI tractare sed ui et uiă peruolgata patrum. (c) Phorm. 439 dicam tibi impingam grandem. Dixi, Demipho. (e) Ad. Prol. 4 indicio de se ipse erit, uos eritis iudices. I think these are all that remain outstanding after the rest have been referred to their several categories already treated: Andr. 753 rogo-no reason why it should be lengthened (see p. 79). Eun. 78 et illas (see p. 73). Eun. 160 et istam (see pp. 72, 73). The probable cure is to read id istam; but the pause might justify et istam. Haut. 74 in illis exercendis. Here DB indicate a variant: they have, teste Umpfenbach, si sumas illis in exercendis. Read si insumas illis exercendis. Haut. 159 et illum. Read Menedeme, et porro recte spero et illum tibi (cf. p. 73). Haut. 437 in illum (cf. p. 73). Phorm. 126 et illos (p. 73). Phorm. 143 ego hinc (p. 73). Hec. 107 ut hoc proferam, sed ut (see p. 73), and add Diomedes' statement (Keil I., p. 437) that the Romans punctuated before sed. Ad. 40 sed ex fratre-the same will hold. Ad. 787 ita-ŭt-dixi. Either a fully coalesced group (cf. ita-me-di-ament) or the opening of a new sentence; punctuate after sunt. Few as these are, the list is still subject to abatements. In (c) it is the special emphasis of TIBI, which shortens impingam (cf. quid înterea, ego interea). Phormio's words are: 'It will be your turn for a swingeing action, Demipho.' In (e) the reading is doubtful. Bentley gave: indicio de sese ipse erit, uos iudices. This leaves us only three instances: (a) ex Graecis bonis Latinas fecit non bonas (Eun. 8). Of the outstanding exceptions, this alone comes from a Prologue. Surprising as it is, a quotation by St. Jerome (in Umpfenbach) establishes it for a current reading as early as Donatus. I suppose it was a Plautine or, at least, a preTerentian tag, quoted here from some earlier Prologue. The taunt was an extremely obvious one for rival playwrights to throw at each other in a literature of adaptations from the Greek. Indeed, one can hardly imagine that it would be left for Terence to say first. Of course the high emphasis on BONIS is a cause, just as in nam expedit вONAS esse uobis (Haut. 388). (b) ui et uiă peruolgata patrum (Haut. 101). uiă-peruolgata may be regarded as a group-i.e. a six-syllable word. Perhaps also archaic, as the alliteration suggests. (d) Read uosmet uidete iam, Laches, Philumena. The plural verb with the singular vocative puzzled some copyist, forgetful of his uestras, Eure, domos. Philumena's name is required. NON-SENARII, non-initial. Finally we come to the disyllables where shortening takes place without initial intensity: 237 'quid istuc' inquam 'ornatist? Quoniam miser quod habui perdidi.' 260 ille ubi miser famelicus uidět mihi esse tantum honorem. 263 si potis est, tanquam philosophorum habent disciplinae ex ipsis. 282 ad illam. PA. Age modo: i nunc: tibi patent fores hae quia istam ducis. 284 qui mihi nunc uno digitulo forès-aperis fortunatus. 943 pro-deúm-fiděm! facinus foedum . . .1 2 1045 illumne qui mini dedit-consilium Troch. Haut. 176 et illam simul cum nuntio ... Troch. Phorm. 516 idem hic tibi quod boni promeritus fueris Hec. ... 594 dum aetatis tempus tulit, perfuncta . . . 173 o facinus indignum! Geminabit ni caues. Ei misero mihi! 180 responde. AE. Ante aedes non fecisse erit melius hic conuicium. 618 nam ut forte hinc ad obstetricem erăt missa. A score of examples out of the six plays; and of this small total no less than one-quarter occurs in a single scene, the dialogue between Parmeno and Gnatho in Eunuchus 232-291, an episode of strikingly Plautine (or at least archaic) colouring, in the broadest, most popular, most motoria of Terence's pieces. The importance of this point is obvious. So Terence has only about a dozen examples of abridgement in an initial disyllabic word, whereas one play (Eunuchus) alone can show as many unabridged. Now, as a specimen of Plautus' practice take the four plays Miles, Most., Persa, Pseud. I find in these only eighteen true cases of iambic length maintained in the initial disyllable of a sentence. 1 I see no cause that should abridge fidem here except the verse stresses on deúm and facinus, which means that it is an unskilful verse-or corruptly reported. Perhaps we should add one to the four recognized troch. quaternarii of Terence (And. 246, Phorm. 183, Ad. 158, 616) and read, Proh deum atque hominum fide> as in And. 246, making a senarius of <O> facinus foedum! O infelicem adulescentulum ! In the example from Phormio the quaternarius tands between two iambic lines. 2 Cf. Caecilius' capit-consilium. 3 Comparing this with Phorm. 949, inepti uestra puerili sententia,' the reading itidem illae mulieres sunt ferme puerili sententia (D1). may commend itself; but, on the other hand, Pacuvius' (Ribb., p. 86) breui capite, ceruice anguina, aspectu truci, looks as if the ablative phrase representing a (Greek) compound adjective was pronounced as a group-word. I.e., eliminating enclitics (Pers. 733, Ps. 545. 770), groups (Mil. 594, Most. 639), the word ehem, which is a mere screatus indefinitely producible, and a dabō (Ps. 118) caused by a full-stop following. That is to say, with Plautus strict (theoretical) iambus length is the exception; with Terence colloquial (slurred) abridgement is the exception-and quite rare at that. But where the matter of abridgement is not a disyllabic word, but a (theoretically) iambic group in a phrase like quis hic loquitur? he regularly made his characters speak the phrase as it was spoken by the 'best people,' no less than by the slurrers and gabblers. In all the six prologues there is no instance of a disyllable abridged, except the famous Eun. 8 (see p. 81). There are only id isti, id esse, quod illi, ita ut facere (init.), magistratus, et in deterrendo (init.), ego interea. If the facts have been here correctly arranged and the inference correctly made, we shall recognize in lines (such as are detailed on p. 82) like Phorm. 342, prior bibas, prior decumbas; cena dubia adponitur, an archaic flavour, just as-even without the hint from Servius, which is not always forthcoming-we recognize a versus Ennianus or a hemistichion Ennianum in Virgil's hexameters. THE UNIVERSITY, GLASGOW. JOHN S. PHILlimore. (I have to thank Mr. R. G. Austin and Mr. C. J. Fordyce, both of Balliol College, for assistance in preparing this paper for the press.) A NOTE ON THE FIRST SALLUSTIAN SVASORIA. 6 IN discussing the authorship of the first suasoria preserved in Cod. Vat. Lat. 38641 I said that an argument against its Sallustian origin had been found in the words 'paulo ante hoc bellum' of 4, 1. By this phrase the author marks an interval of twenty-seven years, and I suggested, as had been done before, that perhaps this is hardly the way in which a man still under forty would refer to so long an interval which had ended only four years before he was writing.' The elasticity of Latin words indicating number and magnitude is familiar, but by the kindness of Dr. Rice Holmes I have been reminded of a place where some of the more interesting examples have been collected. Dr. Postgate, on p. lxxiv of his edition of Lucan VIII., mentions among other Caesarian passages B.C. 3, 106, 1, where the few days concerned in Caesar paucos dies in Asia moratus' are something like three weeks; and B.G. 3, 20, 1, where 'paucis ante annis' dates an event which probably happened either twenty-one or twenty-two years before the time with which the narrative is concerned.2 To these Dr. Holmes would add Cic. in Cat. 3, 1, 3, and, by way of contrast, Caesar B.C. 3, 25, I. In the former the eruption' of Catiline on the night of November 8-9 is described by Cicero on December 3, twenty-four days later, as having occurred 'paucis ante diebus': in the latter a period of scarcely three months is 'multi menses.' It may be worth while to put together such passages in the undoubted works of Sallust as might be quoted in support of the peccant words from the First Suasoria. 'Paucos post annos' in B. Iug. 9, 4 might be an interesting case if it were possible to believe that Sallust had any clear idea of the chronology he was trying to describe; but, in spite of the attempts at defence which have been made, 'nouissume rediens Numantia' (10, 2) seems hopelessly inaccurate, and the result is that the meaning of 'paucos post annos' in 9, 4 remains vague. The interval may indeed be the fifteen years from 133 to 118, as some of the more recent critics have been inclined to think; 1 C.Q. 1923, pp. 160 and 162. 2 For this vide C. Jullian, Histoire de la Gaule (ed. 2), III., p. 107 n. 9, with references there given. 84 A NOTE ON THE FIRST SALLUSTIAN SVASORIA but it may equally well be no more than the last three years of Micipsa's life (11, 6). There is nothing remarkable about B. Iug. 24, I and 86, 4, where the length of time required for a journey from Cirta to Rome and from Rome to Utica respectively is 'pauci dies.' More noteworthy are B. Iug. 41, 1 and 42, 1. In the latter' paucos post annos' means after rather more than eleven years; in the former 'paucis ante annis carries back to a point at least thirty-six years earlier, as is shown both by the following section and by Hist. 1, 11-12 (Maurenbrecher).1 To the words of the First Suasoria with which this note began the nearest approach to a parallel is to be found in B. Iug. 81, 1. There a remark alleged to have been made by Jugurtha in a year which according to Meinel's dating is 107 B.C. is reported as follows-'tum sese, paulo ante Carthaginiensis, item regem Persen, post uti quisque opulentissumus uideatur, ita Romanis hostem fore.' Since Carthage was destroyed in 146, 'paulo ante' here is used to cover an interval of not less than thirty-nine years—an interval longer than that over which the same phrase stretches in the First Suasoria. There is this, however, to be remembered, that the more distant a period of time is from the date of the composition in which reference to it is made, the easier is it to call that period short. In ad Caes. 1, 4, 1, if the document is Sallustian, the end of the interval spanned by 'paulo ante' was only four years before the time of writing, whereas in the case of B. Iug. 81, I it was something like sixty-seven years ago. The effect of these considerations seems to be definitely to weaken the case against the First Suasoria, so that now there is less to set against the evidence in its favour, which to me appears almost conclusive. 11, 9-10 (Dietsch). HUGH LAST. VIRGIL'S CVLEX. MAY I call the attention of English scholars to a remark by Professor Heinze in his review of Professor Frank's Virgil, a Biography, viz. that the Culex was a favourite present for schoolboys in Martial's time (Mart. 14, 185)? How all the difficulties vanish1 if we regard it as Virgil's first publication, a mere tale for a schoolboy, written to help young Octavian in the Greek Mythology class-work! A peg on which to hang this memoria technica had been, we may suppose, supplied to Virgil by a recent incident in Octavian's neighbourhood, the wonderful escape of a goatherd from a serpent. Heinze rightly refuses to believe that the Culex and the Civis can, both of them, come from the same author, at least at the same time. It is easier to believe that the Ciris is Gallus' Epyllium and that Virgil had a hand in its composition. One more remark seems called for. Lucan's reference to the Culex (rightly explained by Professor Phillimore 2), ' and my age is but a step removed from the age at which the Culex was begun,' is not prose but verse : et quantum mihi restat ad Culicem ? It was a part of Lucan's prologue at his first recitatio. 1 Just as all the difficulties in the last couplet of Ecl. IV. vanish if we regard it as a nursery W. M. LINDSAY. rhyme which Virgil weaves into his poem. THE MANUSCRIPTS OF THE METAMORPHOSES OF APVLEIVS. II. In my previous' article I argued that certain of the later MSS. of the Metamorphoses of Apuleius, and especially those which I call Class I., are to a large extent descended from a lost copy of F, made before the rent was torn in Book VIII., and therefore before the writing of ; and I inferred that it was likely that the evidence of these MSS. would prove a valuable addition to that of 6 in places where F is now illegible. In the following pages I shall prove that this hope has been fulfilled. As a necessary preliminary I must say something of the affinities and relative values of the eight texts which I assign to Class I.-seven MSS. and the editio princeps. These fall into two well-marked groups: (a) AI, BI, L1, V2. (b) E, S, N4, a The fact that these eight texts form a single class has already been demonstrated for part of Book VIII., and will appear at large in the course of this article. I have collected many instances of their agreement in omissions, but want of space compels me to suppress this evidence. I have already shown that Class I. has definitely felt the influence of 4, and fresh instances will appear; but this influence is obviously secondary. I have not completely collated any member of group (b); but I have collated much of E and a, and I have checked in E and a all the passages where my records of the readings of group (a) led me to re-examine the text of F and at Florence. It is likely, however, that I may have missed some passages where (b) would have been more suggestive than (a); and, in any case, my re-examination of F and was very much less thorough than I could have wished. Lack of time compelled me to cut down to a minimum my list of likely passages, and in some cases I accidentally noticed unrecorded readings of F, which proved that I. (a) readings, which I had not even intended to check in F, were in fact important: for instance, Class I.'s suspiratus in VI. 29 (151, 13), which I had meant to ignore, proved to be an original F variant. I must have missed many more. Of the members of (a) I have completely collated only B1, which is in the British Museum. I did this at a time when I had only a few secondhand notes about AI. When I examined A1, I discovered that B1 was a direct copy of it. I had not time to collate AI completely, but I checked with it the whole of my collation of B1, and I hope that I have not overlooked much of importance. The evidence that Bi is a copy of AI is uniform and overwhelming. In the first place, B1 (as far as I have observed) contains nothing that A1 omits, but omits several things that AI contains. Further, BI repeatedly misunderstands certain of Ar's abbreviations, and it is always Bi that is wrong. I will add a few striking instances of Br's dependence on pene A1: I. 20 (18, 25) mira et paene infecta FEa, mira et forte infecta A1, mira et pene forte infecta B1 LI. VIII. 1 (176, 21) charite F a LI, carithe E, charise A1, chariset B1. i XI. 26 (287, 28) templi FEa, templo A1, templio B1. 1 Class. Quart. XVIII. 27-42. 2 Butler (Apologia 1914) noted the close connexion in Apol. and Flor. between AI, BI, LI, and V2, and stated that these four, and also E, were not much indebted to . |