of a settled and mild régime under Trajan after the uncertain and transition period of Nerva's reign, the first book, and some years later the second book, of Satires was written and published. In these the events of Domitian's reign were fresh in his memory, and the indignation not yet dulled. So many grievances occur to him that the Satires naturally take a concrete form, are crowded with incidents, full of vivid pictures. But after all, to satirise past times has an element of unreality about it; this is partly concealed in the first burst of indignation let loose, but it was hardly possible to keep it up, and accordingly in the third book (vii.-ix.), though these too belong to Ribbeck's earlier class, the character has already evidently changed, and these Satires stand half-way between the vivid and concrete Satires of Books I. and II. and the abstract colourless disquisitions of Books IV. and V. These latter books, then, are merely the completion of a tendency apparent before. To satirise Domitian's reign under Trajan was permissible, though after a time monotonous; to satirise Trajan's reign under Hadrian was not permissible, even if the improved state of things had not taken away some at least of the materials for satire, and accordingly Juvenal, who it must be remembered, too, was growing old and perhaps garrulous, naturally enough combines his old occupation of declamator and rhetorician with his new rôle of satirist, allowing historical allusions, school theses, and literary reminiscences generally, to take the place of personal and contemporary pictures, the result being what Ribbeck somewhat irreverently terms "afternoon sermons." (3) Similar traces of Juvenal's rhetorical training are apparent in all the Satires alike.

Numerous instances which prove this are

I

collected by Bergmüller (Quaestiones Iuvenalianae). select a very few just to illustrate the argument, not to exhaust it. (a) Rhetorical modes of statement or question are found in the following among many other passages: i. 45, 52; v. 12, 51, 59; vii. 27; viii. 83; xiv. 114; xvi. 7, 35. (b) Rhetorical uses of the inferential particle ergo iii. 104, 281; x. 54; xi. 17, 21, 129. (c) Rhetorical uses of nunc: iii. 268; x. 210; xvi. 35; porro: iii. 126; vii. 98; xi. 9; autem: iii. 209; xi. 90; sed: i. 117; x. 232; xi. 136; xiii. 213; quis, quid or quid enim? i. 48; iii. 86, 147, 208; iv. 46, 101; v. 163; vii. 158, 199; viii. 30, 114, 183, 221; x. 4, 141, 185, 302; xi. 38; xiii. 72, 98, 234; xv. 140, etc.

To this it may be added that the earlier, Satires, though not strictly speaking declamations, are still set pieces, dealing each with some one subject, the illustrations being marshalled in regular order, and are in this respect a strong contrast to the less elaborate and more loosely constructed Satires of Horace, where each Satire deals with everything. Another consequence of Juvenal's rhetorical treatment of his subject, viz. the desire for striking collocation or piquant anti-climax, is also shown. in both classes of Satires, resulting in the non-moral point of view which sometimes reveals itself, as when the rich barber excites as much indignation as the abandoned criminal, the horsey noble as the cruel provincial governor, the luxury of Crispinus as his incest and adultery, the Troica of Nero as his mother's murder.

But while there is a distinction between the styles of Books I.-II., and Book III., there is no less difference, though of another sort, between x., xii.-xiv., and the two last Satires. The former, the declamatory Satires

properly so called, were probably written at Rome under Hadrian, and with all respect to Ribbeck's opinion they are masterpieces in their way, but the two last Satires, or at any rate the fifteenth, for xvi. is so short a fragment that we can hardly judge, are of a different sort. Sat. xv. was probably written in exile, and deals merely with a local incident. Its style is certainly not worthy of Juvenal, and it bears unmistakably the signs of senility. The genuineness of both these Satires has been doubted, but not on grounds which are of very great importance.

With regard to Sat. xv. we may notice that its genuineness was never doubted in ancient times, and that it is quoted in particular as Juvenal's by Servius on Verg. Aen. ii. 540, while the fact that in some few MSS. it is transposed with Sat. xvi. is hardly an argument worth considering. This fact is however, so far as I know, the only argument based on anything like external evidence which has been adduced. Those resting on internal evidence, most fully collected by C. Kempf (De Satura xv. Iuvenalis, etc.), rest on the essentially vague and unsatisfactory criticisms of style and language. He assumes that Juvenal if he had written the Satire must have written it with some motive, and because he can find no motive which seems to him satisfactory he declares it to be non-authentic. Again he lays stress on the awkward digressions in lines 13-23 and 62-72; the weakness of line 55; the tautology of lines 22-25, 47-48, 95, 129-130, 134, etc.; the inappropriateness of lines 84-86 and 89-92; the confusion and vagueness of alterius populi in line 39, and altera pars in line 73; the indefiniteness of inde and hinc in lines 48

e

and 51. Lastly, certain misstatements on matters of fact are alleged e.g. Tentyra and Ombi described as neighbours are really a hundred miles apart; Diana, contrary to Juvenal's statement in line 8, is worshipped in Egypt under the name of Bubastis (Ov. Met. ix. 691); Elpenor in Homer (Od. iv. 552) was not turned into a hog with his companions, but broke his neck by falling down a ladder. Every one will see how insufficient these arguments are to support the conclusion, especially as it is admitted on all hands that both in this Satire and the last there are here and there lines which are thoroughly Juvenalian.

With regard to Sat. xvi. the scholiast says, "ista exploditur a plerisque et dicitur non esse Iuvenalis," evidence, however, which we may fairly say is cancelled by the fact that both Priscian (viii. 6, 31) and Servius (on Verg. Aen. i. 16 and ii. 102) quote the Satire as Juvenal's. Among modern scholars Heinrich is perhaps the representative of those who deny its genuineness, for Ribbeck is disposed to ascribe this Satire to what he calls Juvenal the satirist as opposed to Juvenal the declamator. Apart, indeed, from the fragmentary character of the poem, and its comparative weakness of style, there is really no evidence against its authenticity. As far as the contents go, and the references to military affairs and institutions, there is nothing whatever inconsistent with its having been written in Hadrian's reign, and although Gibbon (chap. v. note 64) says that the style and circumstances of the Satire would induce him to believe that it was composed under Severus or Caracalla, he brings forward no reason to support this statement, and in point of fact there is none. With regard

to the question whether Sat. xvi. was left a fragment by Juvenal, or whether the remainder has been lost, there are two important pieces of evidence both tending in the same direction—(1) line 60 ends a quaternion in P., and there is no subscription "Explicit liber quintus" as there is at the end of all the other books, so that the presumption is that some leaves at the end of the archetype have been lost; (2) this is confirmed by the following calculation: the Scholia to lines 129-158 of Sat. vii. in P. and the Codex Sangallensis 870 are missing, from which it may be safely inferred that one page of the archetype was missing, and that each page of the archetype contained thirty lines. Now if we reckon up from Sat. vii. 159 to the end, counting in the lines occupied by the subscription at the end of each book and the superscription at the head of each Satire, we find again that Sat. xvi. 60 would come at the bottom of a page. The Satire was, therefore, no doubt finished by Juvenal, but has been lost from the archetype from which all our existing MSS. are derived.

On the various subjects contained in this part of the Introduction consult for full information-Teuffel, Geschichte der römischen Literatur, pp. 755-761 in 4th ed., and Studien und Charakteristiken zur Literaturgeschichte, pp. 410 foll.; Friedländer, Sittengeschichte Roms, vol. iii., pp. 459-470 in 5th ed.; O. Ribbeck, Der echte und der unechte Juvenal; C. Kempf, De Satura xv. Iuvenalis; Weise, Vindiciae Iuvenalianae; Bergmüller, Quaestiones Iuvenalianae; J. Dürr, Das Leben Juvenals; A. Weidner, D. Iunii Iuvenalis Saturae, 2d ed., 1889; L. Schwabe, Juvenal's Geburtsjahr in the Rheinisches Museum, vol. xl. p. 25, etc.