Page images

auribus, 5 a; 161 scindit sese unda, 5 a; 175 suscepitque ignem, 2 t; 177 corruptam undis, 3t; 177 cerealiaque arma, 3t; 334 and 349 ante aras, 2 t; 349 atque auri, 4 a; 383 convulsae undis, 3 t; 424 molirique arcum, 2 t; 442 iactati undis, 3t; 506 saepta armis, it; 531 atque ubere, 5 a; 537 perque undas, it; 625 ipse hostes, it; 660 atque ossibus, 4a ; 687 atque oscula, 4 a; 743 unde imbrem, 5 a.

2. Illustrations of the second class are such combinations as: 32 maria omnia, 5 a; 98 animam hanc, 4a; 263 bellum ingens, it; 476 curruque haeret, 3t; 626 seque ortum, it; 191 nemora inter, 4 a.

3. Illustrations of the short accented syllable are such as the following: 114 ante oculos, 2 t; 202 revocate animos, 3t; 385 atque Asia, 2 t; 489 easque acies, 2 t.2

Turning to pause-elisions, the question arises, What are we to consider a sufficient pause to mark the distinction between the two classes of elision? The stronger pauses are marked by punctuation in our editions, and viewing the literature as a whole this might be adopted as our general standard. Punctuation is in many respects arbitrary. Ribbeck's Editio Stereotypa, 1903, which we take as the basis of our discussion of Vergil, differs in this respect not only from other editors, but also from his own earlier editions. In the first book of the Aeneid I have noted upwards of seventy-five pause-elisions in which the second syllable of the elision is long. All but twelve of these are marked by punctuation either in the edition of Ribbeck or Heise, and in all these elisions the pause corresponds to the principal caesura, or to one of the two main caesuras of the line. The following will illustrate the character of these pause-elisions:

13 Karthago, Italiam contra Tiberinaque longe.
48 bella gero.

et quisquam numen Iunonis adorat.
96 contigit oppetere ! O Danaum fortissime gentis.

1 Cf. 69, 175, 189, 218, 243, 429, 475, 476, 524, 537, 547, 695, 738.

2 Cf. 57, 65, 125, 347, 464, 511, 567, 705, 743, 744. I have not attempted to enumerate all the instances of these three classes of elision found in this book, but have given the most striking examples.

251 navibus (infandum !) amissis unius ob iram.
303 corda volente deo; in primis regina quietum.

The second syllable of pause-elisions consists of (1) words not accented on the first syllable, as Italiam, amissis, etc.

(2) words which are usually employed as sentence-enclitics, as et, atque, ac, aut, ut, utque, in, 0, haut.2

There are three pause-elisions in which the second syllable is short:

514 laetitiaque metuque : avidi coniungere dextras.
571 auxilio tutos dimittam opibusque iuvabo.
753 immo age, et a prima, dic, hospes, origine nobis.

In the first of these the second syllable of the elision is accented.

Contrasting the two types of elision in which the second syllable is long, we see that when the elision occurs with words which are closely connected in thought, the second syllable is very frequently strongly accented, and is in many cases the accented syllable of a noun, and that this does not occur in the case of pause-elision. In fact, so rarely does the long accented syllable occur as the second syllable of pause-elision, that the total number of such instances in all the verse of the Golden Age is very small, - smaller even than the number of such cases found in the first book of the Aeneid alone when the words forming the elision are closely connected in thought. This difference of treatment in the two types of elision certainly cannot be accidental.

While observation shows that the nature of the second

1 Cf. 28, 30, 35, 41, 49, 63, 90, 101, 117, 119, 134, 151, 158, 180, 186, 193, 207, 238, 244, 246, 248, 258, 276, 295, 298, 301, 323, 344, 380, 387, 389, 396, 406, 414, 424, 423, 434, 445, 447, 458, 478, 486, 519, 520, 526, 540, 542, 554, 564, 566, 577, 591, 614, 622, 627, 647, 653, 655, 658, 660, 662, 666, 669, 672, 684, 704, 714, 727, 739.

2 Cf. Greek oủ. “ Es war also das Verhältniss der Negation zum verbum finitum dasselbe wie das Verhältniss der Präposition,” Delbrück, Syntakt. Forsch.

IV, 147

8 The nature of this accent will be considered on p. 90.

4 In line 16 hic, and in line 405 ille, occur after a vowel with an intervening pause, and in both cases hiatus is used.

word in pause-elision is regularly as I have stated, yet some apparent exceptions and irregularities occur which deserve special attention.

ii, 78 vera' inquit, i t;? 548 genitori. illi, 3 t; 550 morere.' hoc, 2 a; iii, 45 ego. hic (adv.), 3 a; 408 sacrorum, hunc, 4 t; iv, 35 esto, aegram, i t; v, 484 persolvo; hic, 2 a; 535 longaevi hoc, 4 t; 644 Iliadum. hic, 2 a; 681 posuere; udo, 3 t; vi, 43 aditus centum, ostia centum, 5 a; viii, 364 aude, hospes, I t; ix, 333 singultantem ; atro 3 t; 427 me me (adsum ...,1t; 454 Numaque. ingens, 3t;x, 61 redde, oro, It; 703 comitemque, una quem nocte, 3 t; 905 odia; hunc, 3 a; xi, 353 unum, optime regum, 5 a; 664 postremum, aspera virgo, 5 a; xii, 532 solo; hunc, 4 a; Ecl. i, 13 ago: hanc, 3 a; vii, 8 aspicio. ille, 2 a; Georg. ii, 187 dispicere : hoc, 2 a; iii, ioi praecipue, hinc, 2 a.

Horace, Sat. i, 3, 20 vitia ? immo, 3 a; 4 5, 12 ingerere : huc appelle, 2 a; ii, 1, 83 iudiciumque. esto, 2 t; 2, 30 petere! esto, 6 a; 3, 236 possideam : aufer, 6 a; 283 magnum? addens, 2 t; 307 vitio.accipe, 5 a; 7, 72 ego, hercule, 4 a; Carm. iii, 30, 7 Libitinam : usque.

Catullus 9, 1 Verani, omnibus; 14, 19 Suffenum, omnia ; 75, 4 amare, omnia; 114, 3 aucupium, omne; 8, 9 tu quoque, impotens, noli; 13, 1 Fabelle, apud me; 29, 18 Pontica: inde; 61, 171 aspice, intus; 71, 6 odore, ipse; 77, 2 frustra? immo; 5 eripuisti, eheu; 6 vitae, eheu ; 114, 3 prata, arva, 5 a; 115, 5 prata, arva, i t; 62, 5 Hymenaee, Hymen, 3 t.

1 I shall not give further illustrations of the enclitics mentioned on p. 85 (2).

2 Sinilar elisions of inquit are not infrequent. Cf. Aen. ii, 387; v, 348, 353; viii, 439; Georg. iv, 494. There is not a marked pause before inquit; it is a question whether any pause in the reading is to be made in this case. The sense-pause as well as the principal caesura comes after inquit. Furthermore inquit is an unemphatic word, the weak narrative "s'd he.” The second element of pauseelision may be the accented syllable of a word used parenthetically. This usage

very common in the drama. 8 Similar to the use of inquit (cf. footnote above) is the parenthetical use of oro.

4 I shall not cite further illustrations of ille and hic in pause-elision. I have noted the following cases in Horace and Catullus: Horace, Sat, i, 3, 57 illi; i, 9, 41 ille; Carm. iii, 3, 33 illum; Epod. 9, 6 illis; Sat, i, 4, 136 hoc; ii, 3, 152 hoc; Carm. i, 19, 13 hic; Catullus, 100, 3 ille; 27, 7, hic; 29, 9 and 76, 8 haec; 56, 6 hunc; 91, 2 hoc; 100, 3 hoc; 107, 2 hoc.

Lucretius i, 980 hoc pacto sequar atque, oras ubicumque locaris. Persius iii, 7 itane? ocius adsit 5 a; Juvenal vi, 281 ipsa! olim convenerat 3 t; Statius Thcb. iii, 348 vociferans : arma, arma viri, 3 a.

Turning to those cases of pause-elision in which the second syllable is short, we find numerous examples in which the second element of the elision is an accented syllable even of a noun. As examples we may cite : Verg. Aen. V, 483 tibi, Eryx; vi, 344 responso animum ; viii, 450 redduntque, alii; xii, 142 fluviorum, animo; 945 ille, oculis; Ecl. 3, 88 qui te, Polio, amat; 94 parcite, oves; 97 ipse, ubi.1

I have endeavored to cite from the verse of the later republic and the early empire those cases of pause-elision which might seem to be the most exceptional in relation to the accent of the second syllable of the elision. I have noted certain cases in which there is the sense pause, though it is not indicated by punctuation in our texts, and certain other cases in which punctuation is given, but in which a natural rendering of the thought would not make a pause. Let us examine a few lines in illustration of the latter case :

Aen. viii, 364 aude, hospes, I t. In the case of this vocative we have the standard punctuation, but there it no real pause either in the thought or verse. Brugmann (Vergl. Gramm. 1?, § 1043, 953) states that in the primitive Indo-Germanic the vocative was unaccented when it did not stand first in the sentence. While the Latin has preserved many of the characteristics of the Indo-Germanic, it is not necessary to attribute this character of the vocative entirely to that influence, for we see a similar usage in modern languages. When the vocative is unemphatic as here, it is little more than an unaccented pronoun and it is almost as closely associated with its verb. In view of these considerations and the fact that we find elision in the poets in connection with the vocative, Corssen (112, 780) does not seem justified in his statement that elision

1 Cf. Georg. ii, 18; iii, 95; iv, 172, 318; Horace, Sat, i, 2, 30; 5, 71; 6,61; ii, 3, 117, 150, 180, 260; ii, 7, 2. I have noted ten similar cases in Catullus, but I have not attempted to make a complete list of either of the three authors cited.

is impossible before Eruci in Cic. pro Sex. Roscio, 50 Ne tu, Eruci, accusator esses ridiculus.

Aen. iv, 35 esto, aegram nulli quondam flexere mariti. This occurs in a conversational passage, and such passages allow greater freedom in elision and a nearer approach to ordinary speech, yet there is hardly a more marked pause to be observed here than there would be between a verb and its object clause.

Aen. ix, 427 me me (adsum qui feci), in me convertite ferrum. The nature of the thought indicates haste and does not allow a marked pause before adsum (see also p. 108). Experimental phonetics shows that in the reading of poetry haste is indicated rather by the shortening of the pauses than by the shortening of the sounds.

Lucretius i, 980 hoc pacto sequar atque, oras ubicumque locaris. Here the sense-pause and the caesura come before atque, and after atque there is but a slight pause, if any. Here is also to be taken into consideration that the final vowel in atque had but a slight soạnd and hardly counted as a part of the verse-foot.1

Aen. vi, 43 quo lati ducunt aditus centum, ostia centum. This line suggests the difference which exists between the reading of prose and verse. While in prose there would be an appreciable pause before ostia, in rendering it as verse the pause would be reduced, if not utterly disregarded. While the proper rendering of poetry in all languages makes a distinction between poetry and prose, this principle is not to be carried in the Latin to extremes, which would be intolerable if applied to a living language, as, for example, the disregarding of marked sense-pauses in Aen. i, 48.

Catullus 114, 3 aucupium, omne genus piscis, prata, arva ferasque; 115, 5 prata, arva, ingentis silvas vastasque paludes. The character of these elisions is similar to those last mentioned. They occur in poems of mock-heroic spirit and ironical tone, which contain a higher percentage of elisions than even Lucilius. These lines are intended to be read in harmony with their mock-heroic tone in the heroic style and

1 See p. 93, footnote.

« PreviousContinue »