were, at least on one side, of foreign descent. It is at least possible that the great mass of the native population lacked the desire and the means for erecting such monuments as would endure to our day. But even if Frank's estimate be correct, we must remember that the flood of foreigners came to Italy a few at a time, and that they were people of humble station-Frank thinks they were nearly all slaves. They had every inducement to learn the Latin language as soon and as well as they could. In other words, their linguistic situation finds its closest parallel in the United States of America. The situation has been excellently sketched by Hempl 26 as follows: 'The invader comes in peaceful bands of immigrants, ready to do menial service or follow a humble calling. It is usually not to his advantage to herd with those of his own nationality who are seeking like occupation, but rather to go where there are those who do not care to do what he is willing to do and who will pay him well for his labor. He thus mingles among the natives and settles down in their midst. The better he learns to speak their tongue the faster he gets on in the world. His children play with their children and go to school with them. If in their dress or speech they betray their foreign origin, they are ridiculed and derided as 'Dutch' or 'Irish' or whatever it may be. That they take pains to rid themselves of all traces of their alien origin and avoid the speech of their parents is natural. Their children, in fact, take their turn in tantalizing the children of the later immigrant. In this way vast numbers of newcomers may be assimilated year after year and go to make up a large part of the new race, while their language makes practically no impression on that of the country. This is the story of what is going on in all parts of the United States today. I know it all too well, being myself of foreign parentage. That English is to remain the speech of an American race with comparatively little English blood in its veins, there can be no question." 26 T. A. P. A. xxix, 35. There are differences, of course, between American English and British English, but they cannot be traced to the influence of any of the foreign languages spoken in the United States, except in the matter of vocabulary. Differences in syntax and in phonetics between us and other English-speaking communities are due solely to divergent development. In fact, American English is in many respects more conservative than that of the mother country. This is a tempting topic, but I will merely illustrate with a reference to the retention in a large part of the United States of anteconsonantal r and of the relatively close vowel of such words as half, past, rather, chance, basket.27 Of course I do not mean to say that the linguistic situation in imperial Rome was precisely the same as in present day America, but I can find no essential differences. The outstanding social contrast between the two cases is that the newcomers in Rome were chiefly slaves; but, since the Roman slaves performed much the same services as the recent immigrants into the United States, and since advancement in their case also depended upon their adapting themselves to the alien environment, the linguistic results ought to be parallel. Fortunately we do not have to decide on the basis of general considerations alone. If Frank's hypothesis is sound we shall find that vulgar Latin phonetics differ from standard Latin in about the same way as Greek. Let us try a few comparisons. In late vulgar Latin ē and both became close ẹ, while ě remained as open e and i remained as close i. This looks hopeful, since Greek e is today an open e and ʼn has come to have the same sound as ; but unfortunately is not a close e and has not been since about 150 A.D., and short has not been a close e at any point in its history. To make matters worse, Greek in loan words frequently yields vulgar Latin close e and Romance i; e.g. σxédiov became vulgar Latin scędium η η 27 For details, see Grandgent's delightful essay on "Fashion and the Broad A” in The Nation, c. 13-14 = Old and New, pp. 25-30. and Italian schizzo. Evidently Greek influence will not help us to understand this feature of vulgar Latin. ū In vulgar Latin Ŏ remained an open o, ō and ŭ became close o, and u remained as u. In Greek, on the other hand, o and w became identical in quality as early as the middle of the second century B.C., while v was pronounced [y] at the time when Frank thinks that vulgar Latin was in process of formation. The vulgar Latin accent differs from that of standard Latin chiefly in two respects; penultimate vowels followed by mute and liquid are always accented, and a short penultimate vowel immediately preceded by i bears the accent (muliérem). Greek accent could not account for either of these features. It is true that the influence of the Christian clergy introduced such Greek accentuations as María and sofía; but the clergy were educated men, and this phenomenon belongs primarily to high Latin. At any rate, it has no bearing upon Frank's theory. The outstanding peculiarity of the vulgar Latin consonant system is the palatalization of c before e and i, a change for which Greek presents no parallel. Frank's specific purpose in proposing the hypothesis of Greek influence on vulgar Latin was to provide an explanation for the loss of the quantitative distinctions. I argued above against the old assumption that the Latin quantitative system was a learned imitation of Greek. It is disconcerting to find now that Greek influence can be thought of as tending to break down the quantitative system. Frank must tacitly assume that Greek quantities were obscured before that language could cause a similar change in Latin; but if Greek could make the change spontaneously, why not Latin? I shall elsewhere try to show that Frank's theory of late Latin versification is untenable. If we turn our attention to the contrasts between Greek and Latin phonetics, we find that vulgar Latin always assimilated Greek loan words to Latin speech habits, never the reverse.28 Among the changes mentioned above are the loss of aspiration from 0, 4, and x, the substitution of u or i for AtticIonic v, and the change of penultimate a to e, u, or i. Many less drastic changes in the same direction are discussed in Claussen's article. Finally, and most important, the phonetic development of vulgar Latin and of the Romance languages represents a natural outgrowth of tendencies present in the Italic languages from the earliest times. Oscan of the second century B.C. anticipated the vulgar Latin fusion of original ē and ĭ, and of original ō and ŭ, and it was well along the road toward changing ě in hiatus to ĭ. Umbrian of about the same date had already changed ai to e and au to o. Oscan anticipated the extensive anaptyxis of vulgar Latin, and both Oscan and Umbrian suffered much syncope. The Oscan dialect of Bantia united consonantal i with a preceding t or d to form s or z. Umbrian palatalized k before e and i. It would be tedious to list the many other parallels between the early dialects and the Romance languages. I do not mean to imply that Oscan and Umbrian exerted any direct influence upon vulgar Latin and Romance, although just this is possible in some cases; but Meillet 29 has shown that related dialects tend to develop along parallel lines even when removed from contact. At any rate the early foreshadowing of vulgar Latin changes shows that Italic speech might develop thus without any foreign impulse. 28 Some find an exception in the accentuation of Greek loan words. I do not, but I must postpone my discussion until another occasion. 29 Introduction à l'Étude Comparative des Langues Indo-Européennes3, pp. 410-427. II. On a Passage in Vergil, Aeneid, IV, 550-551 BY PROFESSOR MARBURY B. OGLE THE OHIO STATE UNIVERSITY Of those passages in Vergil's Aeneid about the meaning of which critics and editors have shown and still continue to show a difference of opinion, none offers more interest or is of more importance to the understanding of the poet's deeper meaning than the two lines 550-1 of book IV: Non licuit thalami expertem sine crimine vitam These lines form part of Dido's soliloquy after she has made preparations for her death. In her thought she dwells upon the ways of escape that might be open to her, but, as she immediately admits, they are impossible ways: all that is left is death: 547, Quin morere ut merita es, ferroque averte dolorem. Then, with great dramatic fitness and also with the deepest truth, she recalls rather in sorrow than in anger the part which her sister had played in this tragedy of her life and love: I 548-9, Tu lacrimis evicta meis, tu prima furentem say, "with the deepest truth," for, if Anna had been less practical, less gross, and more spiritual, more like Dido, she would have understood and would never have spoken as she did at the moment of Dido's great temptation (31 f.). These words of Anna are now in Dido's mind and to them we must turn, it seems to me, if we are to find the key to the meaning of the phrases non licuit and more ferae on which hinges the interpretation of the two lines. It would be interesting and no doubt profitable to review the explanations of these lines offered by all who have re |