Besides the above we find other periphrases where simple verbs might have been expected, e. g. 1. 11. 13 uacet alterius blandos audire susurros and I. 14. 13 tum mihi cessuros spondent mea gaudia reges (= reges, quae mea gaudia sunt, mihi cedent).

with Ovid.

In estimating the value of the foregoing evidence we must first bear in mind that these infinitive forms are very convenient for Comparison verse, especially for elegiac verse, and so may be expected to occur with some frequency. The best way of ascertaining what deduction is to be made for this is to examine the frequency with which such periphrases occur in other elegiac poets. As before, I will take Ovid for the test'.

Ovid's elegiac poems contain about 6 times as many verses as Propertius, i. e. about 24,000 lines. In these 24,000 lines I find the following verbs denoting desire, &c., and preparation, &c., occurring with the inf. the following number of times: uolo 31, cupio 15, opto 53, libet 6, desidero 13, posco 12, [utinam over 8 times], paro 15, incipio 10, coepi 5, ordior 4 (3 with loqui), experior 4, pergo 2, molior 12, laboro 5, tempto 15, conor 3. This gives the following result: Propertius 64, Ovid 123. If the same proportion were observed,

Ovid would shew 384. To this must be added the fact that the Ovidian examples have not been weeded like the Propertian, and that, if we counted up the instances where the phrase is really otiose, the

1 I have used Le Maire's Index to obtain these results, and I have selected the most characteristic class that I could, that containing possum not being available through deficiencies in the index. It is noteworthy that several of the verbs occur more frequently in the Metamorphoses than in the Elegiac poems, shewing that too much weight must not be assigned to the metrical convenience of the forms. Their prevalence there is probably due to the poem being a history of inchoate and interrupted lives.

2 Not in Propertius with inf.

Propertian preponderance would be still more evident. I do not claim undue importance for these figures, which are from the nature of the case imperfect, and, I doubt not, in some particulars contestable. I only appeal to them to confirm the impressions which every unprejudiced reader must gather from reading the poems himself.

We have thus seen that they occur far less frequently in Ovid than in Propertius; and unprejudiced and experienced readers will probably admit without further demonstration, that they cannot be regarded as wholly or chiefly the effect of metrical necessity'. Another explanation claims to be considered. May not Propertius have been snared, as it were, by certain forms of rhythm and expression? Or, to put it in another way, may not particular phrases and turns of metre have reproduced themselves mechanically in his poems? That this is possible, will be allowed by all who have watched with attention either their own habits of speech and writing or those of others. The persistence of these forms is one of the most interesting phenomena of language. Any one who has observed how some fashionable trick of verbiage or cantphrase of the hour is taken up and introduced on every possible occasion, not only in the streets but even in educated society, will readily recognize the irregular actions of a principle which is the parent of style in the individual and idiom in a nation. That such persistence of phrases is very marked in Propertius, I should be the last to deny; and that it is to be explained as due to weakness of will, I should not venture to assert. It may at least be probably supposed to be

1 Another view is possible, that they are the remains of a certain redundance of expression which we find in early Latin; in which case they will be related to the other archaisms of Propertius. I do not deny that some of them are. But this will only account for a portion. Compare p. cxliii.

one of his intellectual characteristics, on which more anon. But the question still recurs; 'why was it these particular phrases that manifested the persistence?' To this the natural answer seems to be; because the ideas that they embodied and the tendencies which they conveyed were those most characteristic of the writer. To take an example which might perhaps be thought to make against the argument. It is not the truly brave man who is continually saying 'Who's afraid?' It is rather he who can ask Lack of courage. with the young Nelson, 'Fear! I never saw fear! What is it?' So the recurrence of non ego nunc uereor and similar phrases in Propertius, even in default of positive evidence, would have taught us that he was scantly endowed with physical courage. But that evidence is forthcoming. He naively says he will chase the timorous hare and bird and leave the hazardous boar alone (1. c. on II. 7. 14). He fears a night journey from Rome to Tibur as though it were an expedition to the Gallinaria Pinus, Iv. 15 (16). He freely owns that 'from his blood will no soldier spring,' II. 7. 14 and note.

Subject to the cautions which we have pointed out, the argument will stand, and could, if we had more space at our disposal, be still further supported. For example, it might be plausibly contended that the frequent use of the pluperfect and other completed tenses on the one hand and the periphrastic future on the other (vide infr.) is similarly to be explained. The author declines the effort of contemplating the reality face to face and relegates it, so far as he may, into the buried past or the formless future. Again, the want of connexion in his thought, the frequent change of subject, his eccentric use of particles and other peculiarities which will be discussed below, all point to a weak and unbalanced mind.

We may easily understand how such a mind was influenced by the foreign superstitions Superstition. which had even then begun to infect and

supplant the religion of Rome. Evidently he had dabbled in 'Babylonian calculations.' His allusions to divining and astrology are frequent. And, although he sometimes rises above himself and ventures to jest upon these formidable subjects (as in v. 1), his language more frequently betrays a genuine apprehension of their power. See III. 23 (19), and II. 4. 15 (25) nam cui non ego sum fallaci praemia uati? quae mea non decies somnia uersat anus? v. 5. 9 sqq., &c.

CHAPTER II.

WORKS AND STYLE.

§i Arrangement and subjects of his poems.

Divisions into

THE poems which have come down to us under the name of Propertius consist of 4046 lines of elegiac verse; and of these all but a books. very trifling proportion are genuine. They are divided in all the MSS. which mark such divisions into four books, 1. containing 708, II. 1402, III. 988 and Iv. 948 lines. The unusual length of Book II. has thrown suspicion on the arrangement, and Lachmann and others have divided it into two, II. 1-9 (354 lines), and II. 10-end (1048 lines). In support of this view it is argued that the poems, as we have them, are incomplete. The evidence Assumed inof this is (a) external and (b) internal. completeness of (a) The grammarian Fulgentius, c. 22, ascribes the following line to Propertius, diuidias mentis conficit omnis amor. His authority is however considerably weakened by his also attributing to him a line from an old Latin comedy, catillata geris uadimonia publicum prostibulum: and the line in question has generally been given to Petronius. I may observe that conficio does not occur in Propertius; and if we assign the line to him, we shall have to add diuidiae to his numerous list of archaisms (p. xc.). Servius on Virg. Ecl. 5. 21 quotes testes sunt sidera nobis. This is al

Book II.