tion) "with a Roman nose" (permanent). Nor, as regards mental attributes, can Caesar, B.G. 1, 47, 4, summa virtute adulescentem, have meant that the young man had merely screwed up his courage for the moment.

3. The Genitive was at hand in Plautus's time as well as the Ablative. If in their very nature the Genitive case was fitted to express that which was permanent, and the Ablative that which was transitory, then the Romans would not have begun by putting permanent, as well as transitory, attributes into the Ablative.

Before I pass to my own views, another somewhat prevalent theory of the Genitive deserves a moment's attention; namely, that it originally indicated the possession of some person by a quality. It does not seem to me probable that such an example as magnae virtutis homo meant originally, as Bennett, in the Appendix to his Latin Grammar, § 322, translates it, "virtue's man.” With such an origin, there would be no need of the regularly accompanying modifier. But a more seriously unattractive side of the derivation is that it posits too vague a conception at the outset. Roman thought was much less abstract than

this. A concrete starting-point is necessary for any satisfactory solution.

The theory which I have to propose is as follows:

66

1. The Genitive construction is due to the fusion, more or less complete, of two constructions, the Genitive of Possession, as in such common Roman phrases as "men of the senatorial order," "men of the Greek race," and the explanatory Genitive, as in " a fleet of a hundred ships," an interval of five days." With the shift of the meaning of genus from "stock" or "kin" to "kind," there would grow up a feeling that the case described, and it would then be used with words with which, in the beginning, it could not have been used. Similarly one would not stop with such a phrase as “an interval of five days,” but would, by a very natural association, go on to say a delay of five days," etc., etc. But by this time the effect of the case, on this side also, would clearly be to describe. The range of the construction as a whole would now cover nouns of abstract or general meaning, like class, kind, virtue, and nouns of measure, like mile, foot, year.1

66

2. The Ablative construction is likewise the result of a more or less complete fusion of three constructions: namely, the Ablative of Accompaniment, illustrated by the use of a Preposition in Lael. Schol. Bob.: is cum illo animo atque ingenio; Liv. 32, 9, 3: agnum cum duobus capitibus natum, etc.; the Locative Ablative of Situation or Mental Condition, illustrated by the use of a Preposition in Cic. Sest. 50, 106: in eo statu est; Hor. Ep. 2, 2, 12: meo sum pauper in aere; Cic. Att. 6, 2, 6: magna in spe sum; etc.; and (for the rare eo genere) the Separative Ablative, illustrated by the use of a Preposition in Cic. Font. 19, 42: ex eo genere homines. Cf. Cic. Har. Resp. 28, 61: ut meliore simus statu; Cat. 2, 2, 4: reliquit quos homines, quanto aere alieno; Fam. 12, 28, 3: sum spe bona.

The theory now stated accounts for the facts of the actual uses of the constructions, and the limitations upon each, as follows:

The oldest expression of a mental trait was through the idea of Accompaniment, as in magna virtute. The Possessive idea, as such, was impossible. A man does does not "belong to "a trait. When, however, through the influence of phrases originally Possessive, like eius generis, the Genitive had developed a descriptive

1 Greek has both these constructions of the Genitive, but the free development on the Posses. sive side was arrested.

power, it was then possible to say magnae virtutis homo, and the two constructions were now, for this class of ideas, interchangeable. Even genus itself came to be used, though very rarely, in the Ablative. On the other hand, the phrases eius modi, etc., always maintained themselves unbroken, partly because their extreme commonness naturally gave them permanency, and partly also, doubtless, because phrases like co modo were already appropriated for an adverbial force.

With numerals, the Genitive, originally one of explanation or More Exact Definition, always remained the only possible case. Such a conception as, eg.,

66

a ditch with three feet," was impossible.

Words denoting parts of the body could originally, of course, be used only in the construction of Accompaniment. We may think of a man as "with a Roman nose"; but never of a man as "belonging to a Roman nose." Facies and species naturally followed the same construction (cf. "qua faciest?" "Macilento ore, naso acuto," etc., Plaut. Capt. 646). These words always suggested, in summary, physical details for which the construction would have to be in the Ablative. This, and not the fact that they are of the Fifth Declension, is the reason why facies and species are not used in the Genitive construction until very late. On the other hand, words like statura, forma, figura, tend in a larger degree to suggest the idea of kind (as in homines tantulae staturae, men of such slight stature," "such puny men") and accordingly came to be used occasionally in the Genitive, though the Ablative always remained the commoner construction in classical usage.

66

The origins assigned above also account for the necessary presence of a modifier in either construction. A phrase like "a man belonging to a class" would mean nothing. One would at once ask, "belonging to what class ?" Similarly, one would not say, "a man with a nose." All men normally are equipped with noses, and what one wishes in a given case to learn is with what kind of a nose this particular man is equipped. Similarly, one would have no occasion to say a ditch of feet," but would often wish to say "of such or such a number of feet."

66

The practical results may be summed up in two statements of usage for Classical Prose Latin, and two Notes.

I. Kind and Measure may be expressed by the Genitive.

2. Kind and External Appearance may be expressed by the Ablative; also, in

a few phrases, Situation and Mental Condition.

a. Genus is rarely used in the Ablative construction, and modus never.

b. A few words of External Appearance of a general kind (statura, forma, figura) are occasionally used in the Genitive construction.

30. The Technique of Literary Characterization in Dionysius of Halicarnassus, by Professor George L. Hendrickson, of the University of Chicago (read in the absence of the author by Professor Capps).

The purpose of the paper was to show that Dionysius (or predecessors) had reduced to rule the points of view from which any given literary personality was to be judged. Several methods of characterization were distinguished and illustrated, and their component elements analyzed and discussed. The relation of

the technique of criticism in Dionysius to the other ancient criticism Quintilian, pseudo-Longinus, and Hermogenes - was touched on briefly.

Cicero,

This paper is to be published in the University of Chicago Studies. Remarks were made on the paper by Professor Gudeman.

31. Some Uses of the Prepositions in Horace,' by Professor John C. Rolfe, of the University of Michigan (read in abstract, in the author's absence, by Professor Tarbell).

The discussion of the form of the preposition in Horace was preceded by a general survey of the use of the forms a, ab, and abs, from the earliest to the latest times. It was pointed out that the treatment in our grammars and handbooks is unsatisfactory, in that the differences to be observed in the inscriptions as compared with the literature, and in the various periods, styles, and writers, were not sufficiently regarded. It was shown that the use of ab before consonants was especially persistent in certain stereotyped formulas, such as ab Jove, ab dis, ab re, and with personal and geographical names. The rule of using ab only before vowels and was a gradual development, perfected first in poetry, and appearing in prose first in the writings of Seneca the Rhetorician. Horace belongs in the same class with Virgil and Lucretius in this respect, in contrast to Catullus, Tibullus, Propertius, and Ovid. He has twenty cases of a before consonants, and six of ab, most of the latter being of a formulaic character. The word order of the poets differs essentially from that of the prose writers; Horace offers no special peculiarities. As regards the syntax, Horace has little or nothing that is irregular or peculiar. The readings ab labore in Epod. 17, 24; ab avaritia in Serm. 1, 4, 26; at ipsis saturnalibus in Serm. 2, 3, 4, were argued for against Keller and Holder. The various syntactical uses of ab with the ablative were discussed in some detail, and illustrated as far as possible by citations from Horace.

32.

Tibullus as a Poet of Nature, by Professor Karl P. Harrington, of the University of Maine (read by title).

The Roman poets as a class were not nature poets in any proper sense of the term. This fact seems to me the more remarkable when we consider the natural

charm of Italian scenery. While of the elegiac poets as a class something more in this line might have been expected, because they were rather introspective, and the poets of their own moods and tenses, none among these poets seems on a priori grounds so likely to exhibit a considerable amount of nature painting as the gentle-spirited, war-hating, leisure-loving Tibullus, who was always happiest on his country estate, with his sheep and oxen, the quiet hills and plains, and the starry heavens, all about him.

A search to discover what Tibullus saw, what he loved to see, and what it meant to him, is, however, somewhat disappointing. Although a few passages appear to betray some love of nature's beauty for its own sake, even these are rather monotonous and empty of real feeling; e.g. I, 1, 27; I, 1, 48; 1, 2, 71; 2, 1, 37.

1 This paper will be published in full in vol. xii of the Harv. Studies in Class. Phil.

There is an even larger preponderance than might have been expected of passages referring to the vegetable and animal life of his own farm; and comparatively little to indicate that he had ever travelled extensively, or that foreign scenery had made any impression upon his mind and his imagination. With the phenomena of the outside world, and often even with those of his own home life, his acquaintance is voiced in the merely conventional phrases of the poets: soft garlands,' 'yellow grain,' 'snow-white sheep,' 'cruel wild beasts,' 'caerulean waves,' and 'the unstable sea.' Summer is largely dependent on the 'Dog-Star'; in the sky Jove's thunderbolts, Aurora's car, and Lucifer's star figure prominently; the 'hard' iron and flint and the gems of Ind have their place; streams are 'rapid' robbers; the winds are 'pitiless'; Olympus and Taurus are the types of mountains, though Tibullus never saw either; valleys are either 'deep' or 'shady'; Night 'yokes her steeds'; and fire is Vulcan's 'ravishing' messenger.

The complete list of references may be arranged as follows:

VEGETABLE LIFE.

I, 1, 7: teneras maturo tempore vites.

1, 1, 8: grandia poma.

1, 1, 9: frugum acervos.

1, 1, 15: flava Ceres . . . corona spicea.

I, I, 27: sub umbra arboris.

1, 3, 45: mella dabant quercus.

1, 3, 61: fert casiam . . . benigna rosis.

1, 3, 66: myrtea serta coma.

1, 4, I: umbrosa tibi contingant tecta.

1, 4, 29: quam cito purpureos.

I, 4. 65: robora tellus

vehet.

colores . . .

alba comas.

uvam... spicas.

1, 10, 47: Pax aluit vites et sucos condidit uvae.

1, 10, 67: spicamque teneto... et pomis . . . ante sinus.

2, 1, 3: dulcisque tuis e cornibus uva pendeat.

2, 1, 19: neu seges eludat messem fallacibus herbis.

2, 1, 38: querna pellere glande famem.

2, 1, 40: exiguam viridi fronde operire domum.

Ceres.

2, 1, 43: tum consita pomus, tum bibit inriguas fertilis hortus aquas.

I, I, 31: agnamve

fetumve capellae desertum oblita matre.

I, I, 33: pecori . . . lupi . . . parcite.

1, 3, 45: ferebant obvia securis ubera lactis oves.

1, 3, 59: passimque vagantes dulce sonant tenui gutture carmen aves. 1, 4, 17: docuit parere leones.

1, 4, 31: quam iacet

missus equus!

1, 4, 35: serpens novus exuit annos. 1, 5, 52: e tectis strix violenta canat.

1, 5, 54: a saevis ossa relicta lupis. 1, 5, 56: aspera turba canum.

1, 7, 8: niveis . . . equis.

1, 7, 17: volitet . . . alba . . . columba.

I, 10, 10: securus varias dux gregis inter oves.

e plena rustica porcus hara.

ipse sectatur oves, at filius agnos.
araturos . .

boves.

I, 10, 26:

1, 10, 41:

1, 10, 46:

2, 1, 67

...

inter agros interque armenta Cupido natus :

2, 2, 14: arat valido rusticus arva bove.

2, 3, 8: steriles . . . boves.

2, 3, 20: rumpere mugitu boves!

2, 3, 42: multa innumera iugera pascat ove.

2, 4, 28: niveam

ovem.

2, 4, 57: indomitis gregibus adflat amores, hippomanes . equae.

2, 5, 14: lubrica exta.

...

2, 5, 25: tum pascebant herbosa Palatia vaccae.

2, 5, 38: niveae candidus agnus ovis.

2, 5, 55: carpite . . . tauri, de septem montibus herbas.