STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world by JSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact support@jstor.org. Notes and Discussions 341 HOKACE CARM. i. 34. 14 hinc apicem rapax fortuna cum stridors acuto sustulit, hie posuisse gaudet. Though all the commentators (beginning with the scholiasts) support the accepted interpretation, there seem to me good grounds for regard- ing cum in this passage as the conjunction — "when(i. e., whenever) fortune takes, etc., she rejoices, etc." Besides giving excellent sense and avoiding the somewhat harsh asyndeton, this view makes it unnecessary to account for sustulit as a gnomic perfect which, considering the context, is a doubtful procedure. It also leaves the attendant circumstance in stridore acuto to be expressed by the simple ablative. This is the usual method in Horace. In fact, as an examination of the whole text shows, he rarely inserts cum in phrases of this character. Moreover, cum with the perfect indicative is the precise idiom for the expression of the ante- cedent iterative action which we have here. Horace makes a fairly frequent use of this construction, often introducing it with simul (or its equivalent), ut, ubi, and si. With cum there are six examples, each a perfect parallel to the cum sustulit of the text — Carm. iii. 5. 29; Epod. 2. 17; Sat. ii. 7. 71; Epp. i. 10. 17, i. 15. 18, ii. 1. 205. The position of cum in our passage has doubtless led to the traditional interpretation, but the same ambiguity exists in other places, especially in Epod. 12. 8 and Sat. ii. 7. 53. A Roman would have felt no difficulty in any of these cases. J. Elmokb Stanford University THE SECONDARY ACCENTUATION OF LATIN WORDS OF THE TYPE OF CONSULUISTI In Class. Phil. II, No. 1, Professor Harkness criticizes a paper of mine on a kindred subject which appeared in Class. Rev. XX (1906), pp. 31 ff. In the course of his criticisms, which deal chiefly with minor points, he censures me for suggesting that Latin words of the type of consuluisti bore a secondary accent on the second syllable (p. 61, n.). He attributes this heterodoxy to an unwillingness to admit any exceptions to my pro- posed law of metrical resolution. I select this particular point for notice, not because it is the one most easily disposed of, but because it affords me an opportunity of producing some further evidence in favor of the pronunciation which Professor Harkness condemns. I suppose any word of the measure of consuluisti to have borne a secondary accent on its second syllable, if other inflexions of the same stem had a principal accent in the same place. Thus the principal accent of consului corresponds in position with the secondary accent of 342 Notes and Discussions consiiluisti. I pass by for the moment the evidence to be found in dramatic verse, and propose a method of testing the question which should appeal to Professor Harkness with special force. It is rightly held by most metrists that there is a deliberately produced coincidence of ictus and accent in the last two feet of the classical hexameter line; and that the special rules observed by the classical poets, by virtue of which some ways of ending the line were allowed, others forbidden, had no other object than the producing of such a coincidence. Here there- fore, if we accept this doctrine, we have a means of testing the accentua- tion of Latin words; and if the Koman poets paid attention not merely to the principal accent of a word, but also to the secondary accent, then we can test by this means the position of the secondary accent in words like consuluisti. Professor Harkness holds that the poets did pay attention to the secondary accent (pp. 52 ff.). He is arguing against Havet, who, with Lucian Miiller and some other metrists, denies this doctrine of an intentionally produced coincidence of ictus and accent at the end of a hexameter. Miiller and Havet suppose that in some of the forbidden endings there is a perfect agreement of ictus and accent, and that there- fore, in those cases, the desire to secure such an agreement cannot have been the reason for the prohibition. Havet selects the ending mors sepelires, which is forbidden, although it exhibits, as he thinks, a perfect agreement of ictus and accent. Professor Harkness replies that the agreement is not perfect in endings of that type, because there was a secondary accent on the first syllable of sepelires, which here begins a thesis. Consequently, he adds, "it seems reasonable to assume that this type of ending is avoided because the first syllable of the thesis would be almost as strongly accented as the arsis" (p. 53). I agree with him that the reason why certain endings were avoided was because they did not permit the desired coincidence of ictus and accent; and that, as is generally admitted, there was a strong secondary accent on the first syllable of sepelires, which syllable was thereby unfitted to begin the fifth thesis, properly unaccented. But how comes it that Professor Hark- ness makes no mention of the fact that verse-closes of the type of prae- cipitauit, exorerentur, are just as strongly forbidden? That is the very type of forbidden verse-close which is selected by Luc. Miiller as afford- ing the best proof that the special rules for the verse-close were not designed to secure agreement of accent and ictus (Res. metr. 2 , p. 242). Miiller, of course, shared the customary belief that, if there was any secondary accent at all in such words, it rested on the first syllable. He considers, therefore, that in endings like praecipitauit there is complete harmony of ictus and accent, and naturally infers that the prohibition of such endings must have been due to some other cause than the desire to secure that harmony. How would Professor Harkness meet that argu- ment? He is on the horns of a dilemma: either he must abandon the Notes and Discussions 343 belief that the reason why any particular ending was avoided was because it would have involved the clash of ictus and accent, or he must show how ictus and accent clashed in the forbidden ending praecipitauit. I do not see how he can do the latter, unless he admits a secondary accent on the second syllable. Poets who, like Lucretius, freely admit endings of the type mors sepelires, also admit, and with equal freedom, endings of the type praecipitauit ; poets who, like Virgil, strictly avoid endings of the first type, also avoid, and with equal strictness, endings of the second type (Munro Lucretius, Introd. to Notes II, pp. 13 f.). Unless, therefore, we deny that there is a designed coincidence of ictus and accent in the last two feet of a hexameter line, the avoidance of such endings as praecipitauit is opposed to the view that the secondary accent lay on the first syllable of such words. (The permitted and the forbidden verse-closes are given by Muller loc. cit., pp. 241, 243; and by Havet Me'triqae*, pp. 53 ff.) But perhaps Professor Harkness would seek to escape from his dilemma in the same way in which he seeks to explain the Plautine stressing commdditaUm. He is unwilling to admit that Plautus violated the prose accent in such cases, and equally unwilling to admit that the prose accentuation was commbditdtem. In these difficult circumstances he suggests that, although the normal prose pronunciation was cbm- moditdtem, there was in Latin what he variously calls a "tendency," or a "general tendency," or a "latent tendency," to "accent the first of two or more short syllables" (p. 52). With a passing protest against the use of the word " tendency" in such a connection, as being unscientific, I must deny that any such tendency existed. It is true that words of the measure of mulierem are normally stressed on the first syllable in Plautus; but words consisting of five short syllables are always stressed on the second syllable, e. g., pu6ritia, Mactdonius (Pseud. 50, 347, 1152, 1162; Trin. 845)— a fact overlooked by Professor Harkness — p. 52, n. 1. But even if such a "tendency" had existed, it could not be used to explain the prohibition of such endings as praecipitauit ; for the tendency must, ex hypothesi, have worked with equal potency against endings of the type poscimus arma, condere gentem, probably the commonest type of all, and certainly one of the most favored. Again Professor Harkness compares the Plautine stressing - -=- ~ - - with the occasional stressing of dactylic words on the middle syllable. The comparison is fallacious : the stressing adu6nient6m is found in any part of the line, the stressing corpdre only in the first foot. To his own condemnation of my suggestion Professor Harkness adds a criticism communicated to him by Professor Kadford (p. 61, n.). It advances two arguments. First, the accentuation commbditdtem is said to be " contrary to the old Latin principle of recession (not to mention the earlier principle of initial accentuation)." I confess I do not quite 344 Notes and Discussions understand what is meant by the "old Latin principle of recession." As for the earlier principle of initial accentuation, commbditdtem is no more contrary to it than commdditas is. So also, that consiilue'runt should have replaced the earlier *c6nsuluerunt, is not harder to believe than that considuit should have replaced *c6nsuluit. Secondly, the accentu- ation commbditdtem is said to be "disproved by the absence of pro- celeusmatics like sed exdneraMs, et 6bs6queretHr." The reasoning is not quite sound. The absence of such proceleusmatics might be due to another cause. But, as a matter of fact, such proceleusmatics occur. Ahlberg gives two examples in his Be procel., namely, id Amphitruoni {Am. 893), eo exdnimatiis (Bacch. 298). They were not common, for obvious reasons. They brought a long accented vowel into the thesis, and they were metrically ambiguous. In conclusion, I claim for investigators in the sphere of language and literature what Professor Harkness seems inclined to deny them — the right to call in question any customary belief whatsoever; to examine the evidence, if any, on which it rests; and to form an independent judg- ment, even though they be constrained to depart from the opinions expressed by the most eminent authorities of the present or the past. Dangan House Charles Exon Galway, Ireland Keply to Professor Exon I do not mention the avoidance of the verse-close - w ~ - =s for the reason stated p. 53, n. 1. This type is avoided by the more careful writers because it affords only a secondary accent for the fifth arsis; but it is, however, used more freely than the other non-classical endings (Winbolt Bat. Hex. Verse, p. 134). Hence there is no dilemma for my theory; but Professor Exon's theory encounters two dilemmas: (1) in spondaic endings the type armentorum is freely admitted, but his theory accents arment&rum (Class. Rev. XX, p. 33, "It seems to be generally true," etc.); (2) ^ ~ ~ -i * is the regular accent in comedy (Class. Rev. XX, p. 159; Trans. XXXIV, p. 68), his theory accents the second syllable. Again, words of five short syllables are not always accented on the second syllable, as Professor Exon claims (Ahlberg Procel., p. 28). His five references to one proper name do not prove the accent of a type. I contrasted the relatively frequent use of - a - - s , - i, ~ with the avoidance of ~ * ~ - -, ~ * ~, and state (p. 64) that the exceptions occur more frequently in connection with the secondary accent. Professor Exon's statement that - a ~ occurs only in the first foot is not correct (Klotz, p. 274). Of the two proceleusmatics mentioned by him one does not occur in the editions of Leo and Lindsay, and the other is at least doubtful (Kamain Bes groupes de mots, § 310). Brown University Albert Granger Harkness