disposal apparently no single or comprehensive designation. As in Varro's allusion to Abuccius, so in this satire of Horace his own writings and those of others in the same style are libelli charactere Luciliano. Nor can it be urged that a technical designation is something which we might expect to find avoided. At all events Horace does not hesitate to refer to comoedia here (45) as he does to mimos in the tenth satire (6) in making comparisons with the style which he is defining.

Again in the discussion of the tenth satire, the word satura, comprehending briefly the literary type under consideration, might advantageously have found a place, though no single passages which point to its need so definitely as genus hoc and versus in the fourth can perhaps be indicated. But in such a transition as is made in vs. 7 from the characterization of Lucilius to the universal demands of satirical writing ("ergo non satis est. . . . est brevitate opus"), a vagueness results which I cannot but suspect was due to Horace's inability to use a technical designation for the literary form which he proceeds to define in respect to its spirit and style. The lack of some such specific word of definition gives to the injunctions which follow a universality of application which is obviously not the poet's intention. I have spoken of Horace's inability to use a specific name, but it will of course be clear that its absence might equally well be due to the purist's aversion to a new word.

Concerning this composition in general the observation should be made that the warmth of Horace's criticism of Lucilius has to do only secondarily or even remotely with Lucilius himself. It has to do with a problem of the present, viz., the definition of the character Lucilianus, not as a matter of historical criticism, but as a norm or standard for present practice. The opponents of Horace had according to the letter a very good case in urging that the revived Lucilian style should follow the example of its inventor: should indulge itself in coarse wit, should admit of the prolixity of affected extemporization, should aim primarily at the biting and the caustic, should intermix Greek words with a colloquial freedom which rhetorical prose had long since forbidden. Against all this Horace raised his protest, and urged that, if the character Lucilianus was to take its 1Cf. also vs. 71: "neque pila libellos," and vs. 101: "afore chartis."

place as a literary form in the present, its founder should not be looked upon as faultless: a genius, to be sure, and originator of a form and style untouched by Greek influence, but not a model for a more finished age; nor would Lucilius himself in their own age have written as he had done in the time which fate had allotted to him.

We see here very concretely and definitely the efforts of Horace, as against the theoretical demands and the practice of some of his contemporaries, to pass beyond the personality of Lucilius; to generalize a type of literature which, taking its origin from Lucilius, shall no longer be defined by traits which represented merely his faults and the limitations of his time; in short, to create out of the character Lucilianus a Roman satire. But as yet a recognized and current name for this revived Lucilianism was lacking. For all the efforts of his own circle of literary friends and associates toward the re-creation of older literary forms Horace finds some characterizing words suggestive of their distinctive labors-the comes libelli of Fundanius, the facta regum pede ter percusso of Pollio, the forte epos of Varius, the Camenae gaudentes rure of Virgil-but for himself he says, with the same looseness of allusion which we noted in the fourth satire, "hoc erat melius quod scribere possem," only suggesting as before a definition of his field by reference to Lucilius -"inventore minor." In fine, the character of Horace's references to his literary form throughout Book i bears the mark either of vagueness, or of definition merely by reference to the manner of his great predecessor. This phenomenon so often recurring does not seem merely a matter of chance or of taste in the choice of expressions.1

1 It is worth noting at this point that in two places Horace alludes to his satirical writing with ludo ("haec ego ludo") and inludo (“ubi quid datur oti inludo chartis"). The only fragment of Lucilius which bears any trace of allusion to his work by title employs the words ludus ac sermones (1039, and Marx's comment). The inferior MSS of Seneca give to the satire on the death of Claudius the title ludus de morte Claudi. For Naevius in ludo see below, p. 135, n. 2.

The natural Greek words to represent the idea of satire are κωμῳδεῖν κωμῳδία "voces ad quodvis fere ludibrii genus significandum usurpatae" (Wachsmuth Corp. poesis ludib. 25). In the well-known and chaotic passage of Joh. Lydus De mag. pop. rom. I, 41, xwμwdeîv is used to describe the spirit of Lucilius' writings, and σaruρikǹ kwuwola of the developed Roman satire. What the σατυρικαὶ κωμῳδίαι of Sulla (Athen. 261 c) were in fact it is quite impossible to say, but it would seem probable that the Greek designation represents an interpretation or conception of some poems of Sulla as Roman satires, from the standpoint of a later time. It is almost inevitable that the Greek grammarians who first gave a learned attention to Lucilius-Laelius

Now at the beginning of Book ii Horace, without circumlocution or apology, starts off at once with a technical designation of the Lucilian style "sunt quibus in satura." Here at length we have a name, and together with the name we discern that some progress has been made toward defining it as a literary genus. Its laws have been abstracted and codified ("ultra legem tendere opus"), and Lucilius, though himself the "inventor," is thought of as having composed to a definite end:

cum est Lucilius ausus

primus in hunc operis componere carmina morem.

Here then, whether due to chance or to the developments of the intervening time, there is an absence of that vagueness of allusion which we noted in Book i. We have passed from the character Lucilianus to the recognition of a literary type. But though the occurrence of satura here and its repetition in the sixth poem (saturis musaque pedestri) would seem to point to the full habilitation of the word, yet as it is associated in the first passage with the technical language of a grammatical definition ("ultra legem"), so we may suspect that it was slow to obtain a place in the recognized literary vocabulary of the language. It is not to be wondered at that for his first book of satires Horace should not have employed as title a word which in the text of the book itself he has apparently not felt free to use. But for the second book it is not so clear, and it struck Porphyrio as peculiar that a work which avowed itself as satire should bear Archelaus, Vectius Philocomus, and doubtless others-should have had occasion to assign his work to some general category of literature, and it is not easy to see where else it can have been placed than under the generic heading κwμdia. Such considerations would, perhaps, account for the early association of Lucilius with old comedy, and they may explain why Horace is at pains to state definitely that his own satirical writings are not meant for the stage-an assurance which would have been apparently superfluous with the title satura, but which would have some manifest motive if the Lucilian style were commonly classed as κωμῳδία

haec ego ludo

quae neque in aede sonent certantia iudice Tarpa
nec redeant iterum atque iterum spectanda theatris.

In contrast with these are the true stage comedies of Fundanius,

arguta meretrice potes Davoque Chremeta
eludente senem comis garrire libellos
unus vivorum.

But from these considerations I draw no positive conclusions that Horace's satires were called either ludi or comoediae. Both observations serve rather to show that a fixed category of satire and a name were still lacking.

the title Sermones: "hos duos libros cum sermonum inscripserit tamen de his sic loquitur quasi satura" (ad ii. 1. 1). There is no evidence that the writings of Juvenal, for instance, were ever called either by their author or by the public anything else than saturae: but Horace chose for both books the title Sermones, and why? Was it because, as we have supposed, he chose to use a specific descriptive name within the larger genus, or was the term saturae not yet sufficiently established to be available? Once again, in the epistle to Florus, Horace refers to his satires, and here also not by their technical, generic name, but descriptively as Bionei sermones, adding as if in explanation of their character as satires the words et sale nigro. It will not do to make too much of the passage, but the fact that the precise and technical designation of his odes and epodes has just preceded (“carmine tu gaudes hic delectatur iambis") may raise the suspicion that the word satura was still a stranger even after the experiment of its employment in Book ii. Even as late as 30 A.D. one may infer from the language of Velleius that the word was not yet thoroughly naturalized. For in his survey of the republican literature in ii. 9 he classifies the names which he reviews under general categories of literary effort: "oratores Scipio Laeliusque, etc., Q. Mucius iuris scientia... clara ingenia in togatis Afranii, in tragoediis Pacuvii atque Accii . . . . .. historiarum auctor Sisenna, Rutilius," etc. But in the midst of this series, following the tragedians, he inserts the Lucilian satire with only these words: "celebre et Lucili nomen fuit." For him, apparently, as for Cicero, Lucilius is an individual and not the representative of a literary genus.2

.

But it may be said, is not all this negative evidence, this argument

1 See the writer's discussion, "Are the Letters of Horace Satires?" Am. Jour. of Phil. XVIII (1897), p. 313.

2A brief note in passing. Festus (ex Verrio) quotes from Naevius in satyra (257 M), "quia nam Saturnium populum pepulisti," a fragment which has generally been connected with the two well-known lines quoted by Cicero De sen. 20: "cedo qui vestram rem publicam," etc. The formula with which Cicero indicates the source of his quotation-"sic enim percontantur in Naevii poetae ludo"-has caused much debate, though in fact there is no good reason for questioning the integrity of the text. Now if the two fragments are from the same source, we have, for the purposes of our inquiry, the curious and noteworthy circumstance that Cicero cites the work with the words "in Naevii ludo," while Verrius, a generation later, cites the same work with Naevius "in satyra.”

from silence, outweighed by the fact that Varro had already discussed the etymology of satura and inquired into its origins? It has been a very general assumption that the various speculations which Diomedes and the Horatian scholia give concerning the origin of the word satura and its application to a form of literature go back for the best part to Varro, and in this confidence a special feeling of security has been entertained concerning the foundations of our knowledge. But let us see. In the article of Diomedes the name of Varro occurs only once: "(satura dicta) a quodam genere farciminis, quod multis rebus refertum saturam dicit Varro vocitatum. Est autem hoc positum in secundo libro Plautinarum quaestionum ‘satura est uva passa,' etc. (G.L. I, 485). Diomedes (and his source) of course is discussing the origin of the name of the literary form, but it will be seen that this passage contains no evidence to show that Varro was doing so. Varro was explaining the word satura as it appeared in some connection in the language of Plautus, whether the Plautus which we now have, or a larger Plautus which he may have recognized. Now the word saturam is found in the Amphitruo (667), where it is applied to Alcumena in the natural and obvious meaning of pregnant, and it is to this passage that Marx refers Varro's gloss (Proleg., p. xi). But I cannot entertain so much confidence in a matter so slightly reported to us, and indeed some obvious considerations would seem to me to make it unlikely that Marx is right. For if a grammarian offers a gloss or etymological explanation it is presumably to explain a difficulty or obscurity. This is not present in the case of saturam in the Amphitruo. Nor again would it be any explanation of the adjective use there found to point out another substantive use of the word. That the Plautine Questions would scarcely afford occasion to consider satura as the name of a form of literature need not be said. Of what word or phrase Varro's gloss is an explanation we can only conjecture, but not without plausibility. The archaic phrase per saturam is quoted from Laelius, from T. Annius Luscus (against Ti. Gracchus), and it is found in a fragment of Lucilius. In the time of Cicero and Varro it was already obsolete, and it continued so until it was resuscitated by the conscious archaist Sallust. Here was something of the older language which demanded explanation then as it does now.