We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 11 February 2009
It is sufficiently probable that quantitative scansion in Latin, imposed on a language in which accentuation by stress was alone significant originally, not only gave way to the earlier principle in the decline of Latin literature, but scarcely tended to suppress it at any time in common speech and in familiar writing. It is also probable therefore that even in literature dominated by quantity stress-accentuation was not obliterated altogether. In fact the incidences of it, in Vergilian verse at least, seemed so characteristic that in late Antiquity the Vergilian stress-rhythms were apparently copied without any knowledge of the prosody of the hexameter. In modern expositions, the variety of Vergil's rhythms is generally referred, not only to scansion, but also to stress-accentuation: and it has recently been shown that lines in which stress and metrical ictus predominantly coincide are different in quality and expressiveness from lines in which they do not. Since however no general explanation or definition of this difference in quality seemed to have been offered, I lately attempted a simple theory which would account for the different aesthetic values of lines of these two classes, and which would define, if possible, what these different values are.
page 184 note 2 There is good evidence in the history of the clausula: for the references see A. C. Clark, Fontes Prosae Numerosae, bibliography, ad fin., and Shewring, W. H. in C.Q. XXIV.(1930), pp. 164CrossRefGoogle Scholar sqq., and XXV. (1931), pp. 12 sqq.; and for an appreciation of results, His Grace the Archbishop of York, in Proc. Class. Ass. XXVII. (1930), pp. 24–26Google Scholar.
page 184 note 3 By Commodianus, cited by SirNewbolt, Henry, A New Study of English Poetry, pp. 34Google Scholar sqq.; where the beauty and importance of stress-rhythms in Vergil are well recognized.
page 184 note 4 By MrDale, F. R. in C.R. XLIII. (1929), pp. 165, 166Google Scholar. But Professor H. J. Rose informs me that some slight researches of his own, undertaken a score of years ago and never published, indicated that the verse of Vergil was more ‘heterodyned’ (see below) than that of the early poets (Ennius and Lucretius) on the one hand, and of Ovid and Lucan on the other.
page 184 note 5 Latin Teaching, XIII. (1930), pp. 37Google Scholar sqq.
page 185 note 1 A good example, cited by MrDale, F. R. (C.R. XLIII. (1929), p. 166)Google Scholar.
page 185 note 2 How much the freedom of homodyne contributes with the natural pace of dactylic metre to give rapidity to poetry is shown well by a comparison between Aen. I. 81–91 and Aen. I. 712–722, two passages each of eleven lines. In the first (81–91) there are thirty dactyls and forty-two homodyne incidences; and in the second twenty-seven dactyls but only twenty-eight homodyne incidences. Although the numbers of dactyls are nearly the same, the first passage (81–91) is much more rapid, because the homodyne gives freedom to it. The same passage (81–91) is also a very good example of the adaptation of texture by homodyne and heterodyne to sense, to give either freedom, or constriction and conflict, where the sense requires them. Cf. p. 186, note 2.
page 186 note 1 In determining the incidence of stress-accent, I have of course assumed the truth of the law of the penultima. I have tried to pursue a consistent practice—if perhaps to some extent inevitably conventional—according to the canon given by Westaway, F. W., Quantity and Accent in Latin, pp. 67Google Scholar sqq. Plenum opus aleae: but there is some reassurance in the circumstance that the figures of absolute fourth-foot homodyne for the whole Aeneid varied in two computations, in which different texts were used and different conventions followed, by o'45 per cent. only. I have compiled statistics from the works of Vergil (Hirtzel), the Appendix Vergiliana (Ellis: except Aetna, Haupt), and, for comparison, from Lucretius De Rerum Natura I. (Bailey), Catullus, Carm. LXIVGoogle Scholar. (Haupt), Ps.-Tibullus IV. I (Haupt), Ovid, Met. IGoogle Scholar. (Weise), Lucan, Phars. IGoogle Scholar. (Housman), and Flaccus, ValeriusArgon. IGoogle Scholar. (Kramer). I include Ps.-Tibullus IV. I, in spite of the improbability that the poem is the work of Tibullus himself, and of the uncertain merit of the poetry, because the probable date of its composition gives interest to the comparison with Vergil's usage, and because this poem presents rhythmical qualities which are exceptional.
page 186 note 2 The usages of rhythmic texture exhibited by Valerius Flaccus are not unlike the Vergilian. But though they have restraint and power, they lack expressive flexibility: cf. the storm-passages in Verg, . Aen. I. 81–123Google Scholar and in Flacc, Val.. Argon. I. 574–658Google Scholar.
page 187 note 1 Especially in Ecl. IV. (Smith, E. Marion in Classical Journal XXVI. (1930), pp. 141–143)Google Scholar, where the influence of Catullus is to be discerned also in the metre and metrical typology, but scarcely in the fourth-foot texture.
page 187 note 2 By a definite pause I mean the existence or possibility of a punctuation at least as strong as a semicolon, or of the first of a pair of brackets.
page 188 note 1 For the complexity of the Vergilian technique cf. Bowra, C. M., Tradition and Design in the Iliad, p. 66Google Scholar: ‘It (the Homeric hexameter) existed too early for an elaborate structure of lines to be used, such as, for instance, we find in Vergilian hexameters’: and, for a possible parallel to the conception of texture and pattern in Latin verse, cf. Edith Sitwell, Collected Poems (preface): ‘The poems in “Façade” and some of the songs in “Prelude to a Fairy Tale” are technical experiments—studies in the effect that texture has on rhythm, and the effect that varying and elaborate patterns of rhymes and of assonances have upon rhythm.’ Of this theory the suggestions of the present paper are independent.
page 188 note 2 The metrical fabric of this passage has been discussed already. ‘But the most famous and familiar example of this tour de force in blending two passages into one was discovered by Nettleship and published in his “Additional Notes” in the third volume of Conington's Vergil. The opening of the Aeneid recalls in its substance the opening of the Odyssey, ἂνδρα μοι ἔννεπε, μοȗσα, πολύτροπον, ὅς μάλα πολλά but the rhythm and general structure of the first seven lines are taken from the same number of lines at the opening of the Iliad, so that the first two and the last lines of each group are precisely similar in metre’ (Fairclough, H. R., in C.P. XXV. 1930), pp. 45, 46)Google Scholar. In such careful and elaborate passages harmonious effects of stress-incidence also are generally apparent.
In the citations which follow, I indicate fourth-foot homodyne by the sign / and fourth-foot heterodyne by the sign placed in the margin opposite to the line concerned.
page 190 note 1 E.g. Lucan, Phars. I. 564–567Google Scholar: … matremque suus conterruit infans; diraque per populum Cumanae carmina uatis uolgantur. tum, quos sectis Bellona lacertis saeua mouet, cecmere deos, crinemque rotantes sanguineum populis ulularunt tristia Galli.
—where the punctuation within the ‘movement’ is stronger than at the beginning. I have erred in the direction of including Lucan's questionable released movements in the computation.
page 192 note 1 E.g. Ovid, Met. I. 681–692Google Scholar, an effective alternation: generally, however, they tend to seem inorganic or frivolous
page 193 note 1 These passages of Vergil seem especially capable of this rhythmical analysis: Georg. I. 415–423: II. 1–17: 161–176: 458–489: 532–542: IV. 1–12: 51–66: Aen. I. 1–38: 132–141:223–237: II. 88–99: 199–233: III. 1–12: 712–718: IV. 1–19: 365–392: 630–658: 693–705: V. 1–16: VI. 1–27: 83–94: 417–460 (cf. 450–455 with 434–439): 587 594: 841–874: 888–901: VII. 195–208: 213–242: 591–615: 808–817: IX. 1–13: 146–161: X. 1–15: 132–138: 185–193: XI. 346–361: 903–915: XII. 134–141: 812–840: 940–952.
In some of them (cf. my forthcoming article in the Classical Journal) there seems to be a most elaborate scheme of balances and symmetries, sometimes based on released movements culminating in alternations, or fused together by them. It is unlikely that chance will account for all these formations, especially as they seem typical of Vergil alone. In other poets instances as clear as Ovid, Met. 492–507Google Scholar seem to be very rare.
page 194 note 1 Vergil seems to me to have achieved an almost incredible beauty of art by adopting from others and blending in his own work rhythmical principles of alternation and release, which in the work of others appeared in comparative most triviality and insignificance: as indeed he adopted and blended sounds, images, and phrases of his predecessors and contemporaries. But these adoptions were united immediately as formal antitypes to principles of his own vision and method, principles which Professor R. S. Conway has shown to be of ultimate importance and value in Vergil's poetry. ‘One of the principles of composition which I believe had a large share in shaping Vergil's work, though it is almost too simple to be called a principle, is the method of alternation, or contrast between successive parts of a poem’ (Conway, R. S., Vergil's Creative Art, p. 5)Google Scholar. Amant alterna camenae. ‘The riddle of the story, the δέσις is stated plainly in the celestial debate … And at the end comes the solution, the λύσις (ibid., p. 13). The first quotation matches also the alternation of fourth-foot homodyne; and the second seems to express the conflict and reconciliation of a released movement, precisely corresponding to the first lines of the Aeneid.
