NOTES ON THE MANUSCRIPTS AND

THE TEXT.

OWING probably to the favour with which the Fathers (especially St Augustine) regarded Sallust's writings, we have an exceptionally large number of manuscripts, about fifty, I believe. A start was first made in determining in a methodical way the relative value of this mass of manuscripts by C. L. Roth in 1854. Some of the MSS. have a gap of several chapters towards the end of the Jugurtha. Taking this, and some other slighter omissions, for his criteria, Roth divided the MSS. into three classes, thus:-I. Those which omit J. 103. 2 after necessariorum to J. 112. 3, where they resume with pacem vellet. II. Those which have this matter. III. Those which, besides the filling in of this large lacuna, have certain other sentences, omitted by both the preceding classes. These are J. 44. 5 sed neque muniebantur ea... (though the words vary much), J. 21. 4 de controversiis suis iure potius quam bello disceptare, and C. 6. 2 ita...erat (or est).

The MSS. of the first, and of course oldest class, belong to the 10th century; those of the second to the 11th; while those of the third belong to the 14th, and are therefore of not much account.

8 Nothing has suffered so much in the constant copying as the order of words. In an Ashburnham MS. which I have

That the MSS. of the two first classes come (as has been shown for other authors) more or less directly from one and the same manuscript is almost certain. The following is a neat indication of this. In J. 103. 2, a line or two before the big lacuna, the mass of the manuscripts have a senseless feliciter (which is altered in some to a hardly more satisfactory flectitur). According to Kortte's certain conjecture, it is a gloss, put by the writer or reader of the original manuscript, who thus expressed his satisfaction at the turn things were then taking for the Romans, or, more probably, thus solemnly noted the conclusion under the circumstances of the work. (Cp. the title of the Oratio ' ad Caesarem...de republica incip. felicit.')

From what source the missing chapters were supplied in the MSS. of the second class it is impossible to decide. The loss of them cannot have happened at a very early time, for a grammarian of the 5th century quotes from them: they appear again in MSS. of the 11th century. The MSS. which contain them hand them down to us in so peculiarly corrupt a state, and with so many variants, that, if they were supplied by the recovery of the lost sheets of the original MS., these particular sheets must have been written in some very illegible way or have been much injured. It is more likely that another, a very illwritten and corrupt but complete MS. came to light in the 11th century, and that the writers of the inter

looked at I found, for instance, the following version of C. 35. 1 exemplum earum litterarum est scriptum infra Lucius Catilina mittit salutem Quinto Catulo egregia tua fides cognita mihi grata re tribuit fiduciam meae commendationi in meis magnis periculis.

polated MSS. filled in the lacuna from it. Both these things may of course have occurred.

In the critical edition of Dietsch, which appeared in 1859, the classification of the MSS. proposed by Roth was accepted, and the oldest MSS. were given something of the authority they deserved. But, unfortunately, it was found by Jordan and Wirz that the collation of the two or three of the oldest MSS. was far from correct. At the present time, not all the MSS. of the first class have been collated; but Jordan has done a great service in determining which of the oldest MSS., as far as they are known, are of the most importance. The chief result at which he has arrived is, that the oldest MS. we have is also the best, that it is the main foundation on which the text must be built. The accuracy of the recension of this MS., which Jordan gave out in 1866, has, in all material points, been abundantly established by Wirz, who had previously collated it, and by others.

This leading MS., which belongs to the tenth, or perhaps to the ninth century, is preserved at Paris, and is generally known as 'Sorbonianus 500'10. Its sign is P. The merits of this MS. will, I hope, be sufficiently shown by the critical notes. But it must

9 I take the liberty of quoting a sentence from a letter with which Prof. Jordan has favoured me. "Post emissam editionem secundam contuli aliquot codices; interque eos unum quem adhuc nemo totum excusserat, Vaticanum 2904: sed tamen ut inde vix quicquam bonae frugis domum rettulerim."

10 The MS. originally belonged to the library of Richelieu. It is now in the Bibliothèque Nationale (Fonds Latin 16024). The words ante diem in J. 113. 3, owing to mutilation, are the last decipherable words of the MS.

not be supposed that it is a carefully written or consistently correct manuscript. The writer was a somewhat illiterate person, it is clear. He gives us for instance inuentus simulachelli (for iuventus simul ac belli), quem censores senatu (that word added above the line) probri gratia amauerant, and Piso adulescens nobilissimae audaciae, and other things of the same kind. But the MS. from which he copied was no doubt a difficult one to read. As obvious mistakes have not as a rule been chronicled in the notes, a few words may be said on this not unimportant subject". Many slips the writer himself corrected, and others were put right by a somewhat later hand; but many remain. Of these a large number fall under the following heads :-(a) words are sometimes wrongly divided, either in the way of running separate words together, or by splitting up one word into two: (b) words are lengthened by putting in a syllable in the middle: (c) words are shortened by omission of a syllable: (d) wrong terminations are put to words.

11 Of these errors I give a few examples, choosing mainly such as have not been corrected at all or very unsuccessfully:a) wrong division: exulanas (for: ex Sullanis), simulachelli (simul ac belli), nobilissimae (nobilis summae), ideo (id adeo), minoresunt (minores sunt), suam et (suamet), fugere aut (fugerant), illecon sumerat (ille censuerat), formi dolosa (formidulosa)—b) lengthening: astitutiae (astutiae), consultando (consulto), meminimus (Memmius), vectigales (vigiles), proeliarentur (proeliantur)—c) shortening: dissili (dissimili), primum (plurimum), libertate (liberalitate), ples (plebes), obsidet (obsideret)—d) wrong terminations: foret (fore), agere (ageret), rapiebat (rapiebatur), peterentur (peterent) etc., maxumas hostias copias (m. hostium c.), domos suos (d. suas), etc.

The mistakes of the MS. P' (to be mentioned below) are of a similar character, and they suggest perhaps certain conclusions as to the character of the MS. from which they both come. In that MS. it would seem words were not divided from one another, and abbreviations were often used for the terminations of verbs and nouns and also for middle syllables12. It may be added that the constant mistakes between a and u, and sometimes between ƒ and s, show that the original was in cursive character.

13

Another important and peculiar MS. is preserved in the Vatican Library at Rome Vaticanus 3864 (V). It is of about the same date as P. Of Sallust it contains the speeches and letters from the Catiline and Jugurtha, four speeches and two letters from the Histories 13, and also a speech and letter ad Caesarem senem de republica'-both of them undoubtedly spurious. The corrector of P agrees with V in the parts they have in common, and perhaps this second hand (p) took these corrections from V. The variations between P and V are mainly in the order of the words; and, while in the genuine speeches V gives a later spelling (optimus vulgus vult etc.), in the speech and letter to Caesar we find e. g. Illei, ignarei,

12 I gratefully acknowledge here my obligations to P. Klimscha's Sallustianische Miscellen. 1882. He is of opinion that such mistakes as magorum (maiorum), agebat (aiebat), regi (rei), gemebat (hiemabat) point to some parts at least having been dictated.

13 Has V preserved all the speeches of the Histories? Jordan thinks so. The number seems small, but the ethnographical discussions and geographical descriptions would supply a good deal of ornamentation.