Page images

cultu mutavit; ibid. ii. 6. 19, id vocabulum mutat Mosa flumine ; ibid. iii. 44. 10, pacem bello mutari ; ibid. iv. 23. 7, libertos regios et servilia imperia bello mutaverant; ibid. xvi. 12. 6, menses Maius Claudi Julius Germanici vocabulis mutantur ; Justin, v. 5. 4, Alcibiadem ducem Conone mutaret.

permuto : Hor. Odes, ii. 12. 22, Mygdonias opes permutare velis crine Licymniae ; ibid. iii. 1. 47, cur valle permutem Sabina divitias operosiores ; Val. Max. ii. 6. 8, permuto reliquias spiritus mei fine; id. ix. 1. 7, permutare religionem stupro; Sen. Thy. 598, permutat hora ima summis ; Martial, ix. 22. 12, quem permutatum nec Ganymede velis ; id. vi. 93. 7, virus ut hoc alio fallax permutet odore ; Sil. Ital. i. 660, permutare culmina muris; Val. Flacc. V. 424, permutant carbasa bracis.

verto: Hor. Odes, i. 35. 4, vertere funeribus triumphos; id. A.P. 226, vertere seria ludo; Ov. Met. X. 157, nulla tamen alite verti dignatur.

With muto and its compounds the ablative by many scholars is regarded as one of Means or Price. But in view of the wide extent of the sociative function of the ablative in Latin it seems much more natural to treat the above cases as a sociative development.

10. VERBS OF MATING, WEDDING, ETC. In Slavic verbs of marrying, and betrothing, are construed with the instrumental (Miklosich, IV. p. 701), while in Gothic verbs of the same meaning take the dative, which may well represent the instrumental of the parent speech. In Latin we have :

marito : Hor. Epodes, 2. 9, adulta vitium propagine altas maritat populos ; Col. de Re Rust. xi. 79, ulmi vitibus maritantur; ibid. viii. 2. 12, quae (feminae) ternae singulis maribus maritantur; Apuleius, Met. viii. 8, quovis alio felicius maritare (imv.) ; of. Claudian, Rap. Pros. ii. 89, (Zephyrus) glaebas fecundo rore maritat.

maritus : Hor. Odes, iii. 5. 5, milesne Crassi coniuge barbara turpis maritus, basely wedded to barbarian mate'; Ovid, Her. 4. 134, et fas omne facit fratre marita soror.

geminor : I should regard as very probably sociative the following from Horace, A.P. 13, serpentes avibus geminentur, tigribus agni, though Landgraf, Beiträge, p. 19, takes the case as dative.

adulteror : Hor. Epodes, 16. 32, adulteretur columba miluo. For this I see no satisfactory interpretation except to take the ablative as sociative.

pecco : Hor. Odes, i. 33. 9, peccet adultero. This seems to me entirely analogous to the preceding example. In Gothic, gahorinon, 'commit adultery,' takes the dative, which, as already pointed out, may represent uses of the Indo-European instrumental.

II. VERBS OF Wrestling,

In Vedic we find verbs of contending construed with the sociative instrumental (Delbrück, Altindische Syntax, p. 131). In Slavic verbs of wrestling take the same construction (Miklosich, Grammatik der slavischen Sprachen, IV. 701). In Greek verbs of contending take the dative, which, as is well known, in many of its functions represents the Indo-European instrumental. Traces of this construction seem to be preserved in Latin, in Lucan, iii. 503, ignis viridi luctetur robore. In Cic. de Rep. iv. 4, we have cum with the ablative, cum tuo Platone luctari. The two following passages in Horace may also contain ablatives: Odes, i. 1. 15, luctantem Icariis fluctibus Africum; Epp. ii. 1. 74, luctantur funera plaustris.


These appear in Avestan construed with the instrumental (Hübschmann, Casuslehre, p. 255). The only Latin example I have noted is Apuleius, Met. xi. 15, vultum laetiorem candido isto habitu congruentem. Yet in xi. 27, we find the dative, nocturnae imagini congruentem. Cicero also uses the dative. Hence ambiguous examples, such as Gellius, N.A. iii. 3. 3, sermonis Plauto congruentis, are best taken as datives.


par : Sall. Hist. Frag. iv. 14, scalas pares moenium altitudine ; Ovid, Fasti, vi. 804, in qua par facies nobilitate sua. Merkel changes par to pars , while Peter, who retains the reading of the Mss., has the comment, 'par (sc. Anco) nobilitate.' The sense, I think, is features on a par with her rank’; cf. the repeated occurrence of par cum, e.g. Sall. Iug. 14. 9, quem tu parem cum liberis tuis fecisti ; Cic. de. Rep. i. 7.

aeque : aeque cum occurs in Plautus, Asin. 332, ut aeque mecum haec scias, and repeatedly elsewhere. It is therefore possible that the following examples, which are usually classed as Ablatives of Comparison Katà cúveou are really sociatives : Plautus, Amph. 293, nullus est hoc meticulosus aeque ; Curc. 141, qui me in terra aeque fortunatus erit?

It is perhaps unnecessary to add by way of conclusion that the very wide range of the construction we have been considering has remained hitherto unrecognized. Most of our important manuals of Latin grammar omit all reference to the idiom, while the few in which it is recognized at all give no indication of its actual scope.

V. - The Relation of Accent to Elision in Latin Verse, not

including the Drama.



I'r is the purpose of this paper to consider some of the phenomena of elision in Latin poetry. Though elision occurs most frequently in the drama, yet its use can in some respects be better studied in those forms of verse in which it is employed in a more restricted way and conforms to more definite rules. I shall accordingly consider the subject from this point of view. I desire first to show that the sense-pause occurring in elision should be observed in reading Latin and that there is evidence in the structure of the verse that this was the intention of the poet. I employ the word elision to denote such a union of a final vowel (or vowel with final m), with the initial vowel of a following word as gives the value of one verse-syllable. The term syllable is employed in referring to the second element of elision, though technically this is only one part of the verse-syllable. The term pauseelision is used, for the sake of brevity, to denote those cases of elision in which a sense-pause occurs between the vowels forming the elision. I shall take for granted, without reviewing the arguments, that as a rule both vowels in elision are to be sounded. There seems to be sufficient authority to justify this position, and most modern metricians agree in accepting this theory of the pronunciation.2

I desire first to establish the principle which may be briefly stated as follows: The second syllable in pauseelision does not admit a strong sentence-accent. This principle is not like a physical law, or even like a law of phonetics, working in exactly the same way in all cases. To establish the fact that the poet intended to make a distinction between pause-elision and elision in which the elided words are closely connected in thought, it would be only necessary to prove that he showed a marked tendency to treat the two cases differently,

1 The most important passages relating to this subject which are to be found in our ancient authors are cited in Corssen, Aussprache, 112, 771 ff.

2 Kühner, Lat. Gramm. II, 96; Schmidt, Rhythmic and Metric, trans. by J. W. White, 5 ff.; L. Müller, Rei Metr. Summarium, $ 33, 61; Christ, Metrik der Gr. u. Röm.? $ 44, 32; Gleditsch, Metrik der Gr. u. Röm. in I. Müller's Handbuch, II, 3o, § 41, 89; Plessis, Métrique Grèque et Latine, $ 19, 17.

In order to bring out more clearly the general difference in the treatment of the two classes of elision I shall first contrast the usage in the two cases as exemplified by the first book of the Aeneid. I shall then note those cases in the works of Vergil, Horace, and Catullus which seem to be somewhat exceptional in character, and I shall next consider more briefly the usage of other authors. Vergil makes the most varied and effective use of elision, yet Horace in some of his Satires and Catullus in his shorter poems employs it with greater boldness and freedom. Even in Vergil we see a difference in usage between those passages which are conversational in tone and those which are more formal and elevated in style.

In taking up the first book of the Aeneid, I shall first refer to those cases of elision in which the elided words are closely connected in thought and which have an accent on the second syllable of the elision. I shall divide these cases into three classes :

1. Those in which the second syllable of the elision is long and is the accented syllable of a noun.

2. Those in which the second syllable is long and is the accented syllable of some other word than a noun.

3. Those in which the second syllable is short.

1. As illustrations of the first class we may cite the following: 43 evertitque aequora, 5 a;? 95 ante ora, 2 a; 123 inimicum imbrem, 3t; 142 tumida aequora, 5 a; 152 arrectisque

1 The number denotes the foot in which the elision occurs; a denotes that it is in the arsis; t that it is in the first syllable of the thesis; a small number to the right of the letter denotes that it is in the second syllable of the thesis. Arsis is used to denote the strong, or accented, part of the foot.

« PreviousContinue »