Page images

fabrum (carpenters). The worship of Silvanus was held especially sacred by the carpenters, Silvanus being sometimes called dendrophorus, 'the carpenter.' In the Calpurnius Chius inscription we see that Hilario was very closely associated with the worship of Silvanus. In this respect the inscriptions support each other, and lead to the belief that they both refer to the same Hilario.

Furthermore, the date of the Flavius Hilario inscription corroborates this hypothesis. These lustra belonged to the new series of lustra instituted by Domitian in 86 A.D. and occurred at intervals of four years (see Suet. Dom. 4; Censorinus, 18; Statius, S. iv, 2, 60 ff.; and Pliny, N.H. ii, 47). Accordingly Flavius Hilario held office from 146 to 178 A.D. and had not passed away when the inscription was set up by his wife and daughter. It is reasonable to suppose that he lived to the close of the second century A.D., which confirms the belief that the Calpurnius Chius inscription referring to Hilario was erected at that time.

The third defining clause is iunctus sacomari (for sarcomario), 'hard by the public scales,' misread and so misunderstood by both Mommsen and Dessau, who read funclus (C.I.L. XIV, 309 and XIV, 51). For this use of iunctus, cf. Wilmanns, 1724; and for like expressions see Orelli, 2389 and 2417. For the use of collegius as masculine see Orelli, 2413, 4101,4123,4978, and 7186. To the paper a genealogical table was added.

Discussion by Professor Richardson.
Report of the Auditing Committee adopted. Adjourned at 5 P.M.



The meeting was called to order on Thursday, December 28, at 9.30 A.M.

The reading of papers was continued.

[ocr errors]

6. Old Problems in Horace (continued), by Professor J. E. Church, Jr., of the Nevada State University.

On Horace, Carmina, i, 3, 1-8.

In his interpretation of this passage, PAPA. XXXIV (1903), xxii, in which he suggests the insertion of ut after Vergilium, the late Professor Earle raises two objections to the generally accepted theory that this passage is a benediction and a prayer upon which the former is conditioned, on the ground that if this interpretation be the correct one, there is no reason why the first stanza should have been the first and the second the second, — “indeed, it would be a great improvement if the two stanzas were to change places,”. nor is it to be supposed that Horace wrote arrant nonsense here.

We should raise no question against the first objection if these stanzas were the product of the English mind and language. But several examples of Roman benedictions followed by prayers strikingly similar in arrangement and language to the above stanzas cast much doubt upon the tenability of the position taken. These examples, moreover, occur in formal inscriptions as well as in literature. Such are Bücheler, Carm. Lat. Epigr.: 197 Ita levis incumbat terra defuncto tibi ...

rogo ne sepulcri umbras violare audeas;


194 Ita candidatus quod petit fiat tuus

opus hoc praeteri; 195 Ita candidatus fiat honoratus tuus

et ita gratum edat munus tuus munerarius

et tu (sis) felix, scriptor, si hic non scripseri[s]; and CIL. VIII, 1070 Ita tibi contingat quod vis, ut hoc sacrum non violes. The benediction in every case precedes the prayer. The introductory ita is equivalent not to hoc modo, as Professor Bennett has suggested in his note on the ode in question, but to hac condicione, as is shown not only by the general sense of Büch. 197 and CIL. VIII, 1070, but also by the imperative in Büch. 194 and by the conditional clause in Büch. 195. Moreover ita and not sic is apparently the original particle, the choice of sic in Horace, Ovid, Tibullus, quoted below, and in Büch. 215, 3, having been evidently constrained by the metre while ita was chosen without visible constraint in the prose inscription CIL. VIII, 1070, and in Büch. 194 and 195 was employed in evident preference to sic, in spite of the simpler form of the iambic senarius available.

In further defence of this theory, can be cited Büch. 196, 1467, Tibullus i. 4, 1-3, and others.

It is conceded, on the other hand, that the reverse order is also employed, as in Büch. 1458, 1466, Ovid, Amor. i. 13, 3-4, and Tibullus ii. 5, 121-2; iv. 4, 19

The order of the first was determined by the thought expressed, that of the others possibly by the extreme shortness of the prayer which, artistically consid. ered, could not fittingly follow so long a benediction.

The curse, where this stronger means of defence is employed, does not appear to precede the prayer. In fact, the distinctive prayer rarely occurs, the curse appearing in the conditional form, as in Wilmanns' Exempla 271, quisquis huic sepulchro nocere conatus fuerit manes eius (elvs) eum exagitent. An example of a prayer followed by a curse similar to the above may be found in Henzen, 6977.

The second objection may be dismissed as being over-critical. The use of personification found in all literature is but fully applied here, as in Vergil's Aeneid, i. 168, Hic fessas non vincula navis | ulla tenent. The ship is addressed as human with human characteristics. The ship loves not the storms more than does the sailor, and is “ wearied ” as much as he. Therefore, if the ship will guard its precious burden, may the buffeting of the storms be taken away and favoring winds direct its course. The only crux which a critical rather than a poetic mind would see is that the ship needs no such inducement, for the passenger's welfare and its own are identical. But this is really no crux ; the prayer is the natural expression of a solicitous heart. When critics find aught to assail in Tennyson's prayer to the ship which bears the ashes of his friend (In Memor. Canto XVII),

my prayer
Was as the whisper of an air
To breathe thee over lonely seas.


[blocks in formation]

then may

first be cast a stone at one of the most tender and imaginative passages in Horace.

Discussion by Professors Fairclough, Clapp, Schilling, Murray, Foster, and Richardson.

7. On Correption in Hiatus (concluded), by Professor E. B. Clapp, of the University of California.

The study of this subject leads to the following conclusions : 1. The practice of the poets as regards correption is influenced somewhat, though not so much as we should expect, by vocabulary and style. The very frequent occurrence of a given vowel or diphthong at the end of words is not always accompanied by a corresponding frequency of correption. Conscious or unconscious choice must have played its part. 2. As regards the origin of the usage, the consonantization theory of Hartel and Grulich offers too exact and satisfactory an explanation of many of the phenomena to be wholly rejected. 3. If correption in hiatus began with the “short" diphthongs al, el, 06, in accordance with this theory, its origin must go back to forms of poetry older than our Homer, since in the earliest as well as the latest portions of the Iliad and Odyssey we find a tolerably settled and stereotyped practice, and the curtailment of quantity is by no means confined to the diphthongs mentioned. 4. Whatever tendency exists in the later poets toward extending correption beyond the Homeric limits (as to a slight extent in Hesiod, Simonides, Manetho) must be regarded as poetic experiment, in a direction which did not win general approval. 5. The general tendency in the later poets, in this as in so many other features of metrical usage, lies in the direction of the limitation of the poet's freedom, and the setting up of fixed and conventional standards.

This paper is printed in full in Classical Philology, Vol. I, pp. 239252.

Discussion by Professors Murray, Bradley, and Richardson.

8. The Helen Episode in Vergil's Aeneid (ii. 559–621), by Professor H. R. Fairclough, of the Leland Stanford Jr. University.

The writer maintains that 11. 567-588 are genuine. Thilo's objections are first examined and answered. Heinze (Virgils epische Technik, p. 45 ff.) adds other arguments in condemnation of the passage. The words scilicet haec Spartam incolumis, etc., if genuine would furnish the only soliloquy in the narrative of the 2d and 3d books. “ Wie unnatürlich und frostig!” But the soliloquy will appeal to most readers as unusually impressive, and from the artistic standpoint seems to be modelled with great care. Thus Wagner comments on the beautiful balance between the three questions in the simple future, aspiciet, ibit, and videbit, and the three in the future-perfect, occiderit, arserit, and sudarit. Servius had noted that 1. 601

Non tibi Tyndaridis facies invisa Lacaenae refers to the expunged passage in which Helen is introduced, but what, asks Heinze, are we to say about culpatusve Paris (602), of whom there is no mention


in the preceding lines ? But culpatusve Paris is only a corollary to the previous words about Helen. If she can arouse such anger, so also surely can her guilty paramour. The two have the force of a plural. It is no human agents you must

It is the gods themselves who are responsible for Troy's downfall. Heinze's idea that 11. 601-602 would be natural enough apart from a previous passage involving Helen or Paris, is quite alien to the directness of Vergilian narrative, though it may be paralleled in Greek tragedy, especially in lyrical pas. sages. Heinze himself has seen that Vergil probably had in mind here the famous passage in the Iliad (T 164):

ου τί μοι αιτίη εσσί, θεοί νυ μοι αιτιοί είσιν,

οί μοι έφώρμησαν πόλεμον πολύδακρυν 'Αχαιών. Here we have the directness of Epic style. The words are addressed by Priam to Helen. So, too, all is simple and direct in Vergil, is, as we believe, Helen is present in the scene, but how different, if, as Heinze holds, Venus mentions her merely as the ultimate cause of Troy's downfall !

Further, Heinze enlarges on the ancient criticism : turpe est viro forti contra feminam irasci. The mere irasci, he says, would not dishonor Aeneas, but Vergil would never have allowed his pious hero to conceive the thought of killing a defenceless woman, especially if she had sought refuge at the altar. How would this, he asks, befit one who has just narrated with horror the story of an altar-desecration ? But let us remember that the thought is never carried into action, and that the hero himself has anticipated criticism (ll. 583 ff.). That Helen is a nefas (585), an unholy thing, is (at least at such a time) a sufficient defence against the charge of impiety. Heinze's whole argument is an elaborate example of special pleading.

An Homeric situation in many ways similar to this Helen episode is one to which sufficient weight has never been given. Henry calls attention to the resemblance, but has not developed the parallelism. See Odyssey T 1-55. Here, as in Vergil, the hero meditates the slaying of women, but does not carry his thought into action. Here, too, the hero soliloquizes, and here, too, a dea ex machina appears on the scene. In Homer, Athene reminds Odysseus of his home, his wife, and child, and in Vergil Venus reminds Aeneas of his father, his wife, and son, though even closer is the parallel in 562:

subiit deserta Creusa, Et direpta domus et parvi casus Iuli.

And still further, as Athene chides Odysseus for his lack of confidence in divine aid, and assures him of her protection to the last ; so in Vergil, Venus confessa deam, ' manifesting the goddess,' reproves her son, first for his frenzy - quid furis (595)? — but secondly for forgetting her — quonam nostri tibi cura recessit? and the Homeric parallel is sufficient to determine the exact meaning of this rebuke, which does not mean that Venus is wounded in her feelings because Aeneas has thought of attacking Helen, or because she has a special interest in Anchises, but because (as Conington puts it) “ Aeneas by losing self-command showed that he had lost confidence in his mother and sense of his relation to her." Lastly, the goddess assures Aeneas of her unfailing support (1. 620).

The parallel is fairly complete, and the conclusion seems irresistible that as this Homeric scene must have been in the mind of him who composed 11. 567588, as well as of the author of the succeeding lines, the whole of the passage involved, the doubtful and undoubted lines alike, must be the work of one and the same poet, viz. Vergil himself.

From the account of Vergil given by Suetonius we may draw many important inferences. In the first place, a work of such magnitude as the Aeneid, involving the use of a great variety of legendary and historical material, must, if composed bit by bit (particulatim), and in irregular order, have been subject to numerous imperfections and inconsistencies until the work of revision was complete. Hence the inconsistency noted by Servius.

In the second place, parts at least of the Aeneid must have been more or less known before the edition of Varius and Tucca appeared.

In the third place, it was the poet's practice to discuss his doubts and difficulties with others, and doubtless the two to whom he turned most frequently were his two greatest literary friends, Varius and Tucca. These therefore were familiar with the poet's sentiments and conceptions, and though the emperor's commands prevented them from destroying the Aeneid, according to Vergil's express entreaty, yet they were in a position to see that, as far as possible, the poet's wishes should be carried out. Vergil had probably expressed his dissatisfaction with the Helen episode, and his executors decided to omit it. Inasmuch as the emperor's instructions prevented them from making additions, they were compelled to leave the context in an imperfect state.

But the passage was already known to others, and was possibly published later by some one who regretted its omission. Indeed, the very fact of its omission from the first complete edition would bring it into notice.

In lieu of the substitute passage which we may well believe Vergil intended to compose, we are justified in retaining in our texts the one which Servius has preserved, believing that though its author was dissatisfied with it, as indeed he was with the Aeneid as a whole, yet it is the work of Vergil himself, and that the second book suffers vastly more from its omission than from its insertion.

The paper appears in full in Classical Philology, Vol. I, pp. 221-230. Discussion by Professors Johnston, Senger, and Murray.

9. The Yokuts Indian Language of California, by Dr. A. L. Kroeber, of the University of California.

The Yokuts language is notable among American languages for the small number of its affixes and elements used in composition, and its consequent simplicity of structure as regards word-building by synthesis. It totally lacks pronominal incorporation, which is regarded as one of the most important characteristics of American languages in general. Its pronoun, which approximates in function the pronoun of the modern analytical Indo-European languages, is very systematically regular and apparently shows a strong influence of an analogizing tendency. A notable feature of the languages is a complicated system of vocalic changes in the stems of words. These changes appear to be occasioned by suffixes, but are generally not determined by the vocalic content of the suffix. Any particular vowel change is primarily dependent upon the grammatical idea

« PreviousContinue »