:

here definitely quoted from De Sen. c. x. § 33, is as follows:'Cursus est certus aetatis et una via naturae eaque simplex, suaque cuique parti aetatis tempestivitas est data.' This seems to justify the corrections generally made by modern editors, viz.: 'Certo corso ha la nostra età e una via semplice, quella della nostra buona natura: e a ciascuna parte della nostra età è data stagione a certe cose.'

It will be remembered that alla and ha la are practically the same in many MSS.; so also are è and e; and in the case of the e following semplice, its similarity with the final letter of the preceding word makes it almost optional, and a mere matter of editing,' whether we insert it or not. The repetition of buona with età, as well as natura, seems to be a mere blunder.

(17)

Conv. IV. xxvii. ll. 134 seqq. Dice Tullio . . . nel libro degli Officii: 'Sono molti certo desiderosi d'essere apparenti e gloriosi, che tolgono agli altri per dare agli altri; credendosi essere buoni tenuti, se arricchiscono [gli amici] per qual ragione esser voglia. Ma ciò tanto è contrario a quello che fare si conviene, che nulla è più.'

This passage is found in Cicero, de Off. I. c. xiv. § 43:'Sunt autem multi, et quidem cupidi splendoris et gloriae, qui eripiunt aliis quod aliis largiantur: hique arbitrantur se beneficos in suos amicos visum iri, si locupletent eos quacumque ratione. Id autem tantum abest officio, ut nihil magis officio possit esse contrarium.'

All editors seem to agree in inserting 'gli amici' here, partly from the want of an object to arricchire; partly from the previous context, which seems to imply the enrichment of others, rather than oneself, by dishonest means; and still more from the general purpose of the passage of Cicero, which is very closely and accurately followed by Dante, and in which the enrichment of friends is a prominent feature. I do not feel quite confident of the necessity of such an insertion, but it seems fairly reasonable, as the accidental omission of the words by an early copyist might account for their disappearance from existing MSS. The very slight change of certi

to certo seems to be justified by the original 'sunt autem multi,' especially as certo would be almost sure to fall under the attraction of molti, and become certi.

(18)

The direct quotation of Cicero, de Sen. c. xxiii. § 83, in Conv. IV. xxviii. ll. 45 seqq., enables us again to rectify a corruption in the text. The words in Cicero are: Equidem efferor studio patres vestros, quos colui et dilexi, videndi: neque vero eos solum convenire aveo, quos ipse cognovi, sed illos etiam, de quibus audivi et legi et ipse conscripsi.' The lect. vulg. ran thus: A me pare già vedere e levomi in grandissimo studio di vedere li nostri padri che io amai, e non pur quelli, ma eziandio quelli di cui udii parlare.' The two following corrections seem to be justified :—(1) Certainly the change of nostri to vostri (words that are constantly interchanged, and which are nearly indistinguishable in many MSS.); (2) probably the omission of the initial words, 'A me pare già vedere e. For (a) the passage in Cicero pointedly begins with Equidem efferor, &c., without anything at all corresponding to these words. (b) They run awkwardly, not to say inconsistently, with the words that follow, since if Cato already seems to see them, it is rather an anticlimax to add that he eagerly desires to see them. (c) The words in themselves look suspiciously like an 'improvement' of the quotation, to make it more directly apposite to the preceding words which it is adduced to illustrate, viz. vedere le pare coloro,' &c. (3) Previous editors have agreed to insert after 'non pur quelli,' the words 'ch' io stessi conobbi,' to correspond to 'quos ipse cognovi,' but I scarcely feel that we are entitled to do this without further evidence, since the whole of the latter part of the quotation is somewhat curtailed, and these words may have been omitted in that process, since the essential part of the quotation has already been sufficiently given.

(19)

There is a curious misquotation of Cicero, 'nel primo degli Officii,' in Conv. IV. xxv. 1. 95: Nullo atto è laido, che non

sia laido quello nominare.' What Cicero really says is quite different, viz.: 'Quod facere turpe non est, modo occulte, id dicere obscenum est' (I. xxxv. § 127). Dante's memory seems at fault here, since he apparently had a definite recollection as to the source from which he is quoting, otherwise we might have thought that he may have been misled by a passage in the Tesoro of Brunetto Latini, Bk. VII. c. xiv. fin. (ed. Venice, 1533): 'Socrate disse, “ciò che è laido a fare io non credo che sia onesto a dire.”

(20)

From a longish quotation of Cicero, Paradoxa, ch. i. § 6, occurring in Conv. IV. xii. ll. 55 seqq., in which the original is closely followed throughout, it appears as though Dante had read dixi instead of duxi-a case exactly similar to one noticed in reference to Ovid (No. 16). The portions bearing on this point are as follows:-'nè l' allegrezze, delle quali massimamente sono astretti, tra cose buone o desiderabili essere dissi. In Cicero the passage is: neque eas quibus maxime astricti sunt voluptates, in bonis rebus aut expetendis esse duxi.' I can find no evidence of any such variant as dixi recorded.

(21)

There is a curious variation in a quotation of a passage in the De Officiis in De Mon. II. x. 1. 38. Dante quotes Cicero thus: Sed bella, quibus Imperii corona proposita est, minus acerbe gerenda sunt.' The quotation is accurately made from De Off. I. xii. § 38, except in the substitution of the word corona for gloria. I can find no variant recorded in the passages either of Cicero or Dante.

(22)

In Conv. IV. xxiv. ll. 108 seqq., there seems little doubt that the illustration from the vine is derived from the De Senectute, xv. §§ 52, 53, since we have seen that Dante was specially familiar with that work. It is to be observed that the two illustrations by which the' argument from design' in 'la buona Natura' is illustrated, viz. the tendrils to support the branches

of the vine, and the leaves to protect its fruit, occur in both passages, though in inverted order. We may compare the way in which twice in ch. xxviii similes or illustrations have evidently been suggested by the De Senectute-the comparison of death in old age to the dropping of ripe fruit from the bough (1. 28), and that of the soul at death returning to its own home, after living as a stranger at an inn (1. 51). See De Sen. §§ 71 and 84 respectively.

(23)

Finally, we note what seems to be a pseudo-Ciceronian quotation in Conv. IV. xxix. 1. 72 :

E però dice Tullio, che 'l figliuolo del valente uomo dee procurare di rendere al padre buona testimonianza.'

I have been unable to find anything like this anywhere in Cicero, and some friends who have made a special study of Cicero, have been equally unable to identify the quotation. The absence of the usual definite reference to the Book, as well as to the author, is suspicious, and disposes me to think that Dante was probably quoting secondhand from some collection of 'Adagia,' &c., in which this sentiment may have been assigned to Cicero. This certainly seems to explain some doubtful quotations elsewhere, e.g. from Aristotle, in Conv. IV. xi. 1. 83, and from Seneca, in Conv. IV. xii. 1. 1201. There can be no doubt that before the days of printed books, many references and quotations must necessarily have been obtained from such sources, and not directly from the authors cited.

LIVY.

Dante's citations of Livy present many curious features. The references given in the Index, and also some of the characteristics exhibited by the passages discussed in the next few pages, seem to point to the conclusion that Dante's acquaintance with Livy was slight, and probably mostly at

1 See under Aristotle, No. 71, and under Seneca, infra, p. 288.

T

6

second hand. (1) We note that the references, such as they are, come almost entirely from the first few Books. (2) It is very significant that, in the case of Livy, Dante entirely departs from his usual practice of specifying the number of the Book from which he is quoting, as for instance in the cases of Aristotle, Virgil, Ovid, Lucan, Statius, Cicero, &c. In no single case does he do so for Livy. He is usually content to refer to 'Livy' quite generally, but when he attempts in any way to specify the source of the quotation, it is only in such vague expressions as- nella prima parte del suo volume' (Conv. III. xi. 1. 31); in prima parte' (De Mon. II. iv. 1. 33, and xi. 1. 37); in prima parte sui voluminis' (De Mon. II. iii. 1. 34); ' in bello Punico' (De Mon. II. iv. 1. 64). (3) The statements attributed to Livy are in some cases not found there at all, being in fact apocryphal, and in other cases very partially so found. Occasionally the language of Dante corresponds more nearly with what is found in Orosius, or perhaps (as Dr. Schück suggests 2) even in Florus. The general impression left on my mind is that Dante probably used some historical epitome, and either on its authority, or perhaps on grounds of general probability, assumed that Livy would be the natural source from which such information would be derived. This impression is strongly confirmed when we find a pure fable like that connected with Alexander the Great (see infra, p. 278) fathered upon Livy3, in whose works there is no trace of it, so that it is not a question of inaccuracy of memory in mere details.

The following are the passages which seem to call for some special remark.

(1)

The reference to Livy in Inf. xxviii. 12, as the authority for the story of the bushels of rings of the slain Romans collected by Hannibal after the battle of Cannae, seems to explain the epithet alte as applied to spoglie in that passage. Livy (xxiii. 12) speaks of 'annulos aureos, qui tantus acervus

1 See supra, pp. 15, 96.

2 Neue Jahrbücher für Philologie, 1865.

3 See also p. 277, under No. 5.