BOOK I. (continued 2)

§40. Iunctos: how can anything be a compound of one thing? The notion that iunctos could mean aptos (R. and P. 366) is untenable. I entirely agree with Madv. (first Excursus to his D.F.) that we have here an anacoluthon. Cic. meant to say iunctos e quadam impulsione et ex assensu animorum, but having to explain φαντασια was obliged to break off and resume at sed ad haec. The explanation of a Greek term causes a very similar anacoluthon in De Off. I. 153. Schuppe, De Anacoluthis Ciceronianis p. 9, agrees with Madv. For the expression cf. D.F. II. 44 e duplici genere voluptatis coniunctus Ernesti em. cunctos, Dav. punctos, ingeniose ille quidem says Halm, pessime I should say. Φαντασιαν: a full and clear account of Stoic theories of sensation is given by Zeller, ch. V., R. and P. 365 sq. Nos appellemus licet: the same turn of expression occurs D.F. III. 21, IV. 74. Hoc verbum quidem hoc quidem probably ought to be read, see 18. Adsensionem = συγκαταθεσιν. In nobis positam: the usual expression for freedom of the will, cf. II. 37, De Fato, 42, 43 (a very important passage). The actual sensation is involuntary (ακουσιον Sext. Emp. Adv. Math. VIII. 397). Tironum causa I note that the Stoics sometimes speak of the assent of the mind as involuntary, while the καταληπτικη φαντασια compels assent (see II. 38). This is, however, only true of the healthy reason, the unhealthy may refuse assent.

§41. Visis non omnibus: while Epicurus defended the truth of all sensations, Zeno abandoned the weak positions to the sceptic and retired to the inner citadel of the καταληπτικη φαντασια. Declarationem: εναργειαν, a term alike Stoic, Epicurean, and Academic, see n. on II. 17. Earum rerum: only this class of sensations gives correct information of the things lying behind. Ipsum per se: i.e. its whole truth lies in its own εναργεια, which requires no corroboration from without. Comprehendibile: this form has better MSS. authority than the vulg comprehensibile. Goerenz's note on these words is worth reading as a philological curiosity Nos vero, inquit: Halm with Manut. writes inquam. Why change? Atticus answers as in 14, 25, 33. Καταληπτον: strictly the thing which emits the visum is said to be καταληπτον, but, as we shall see in the Lucullus, the sensation and the thing from which it proceeds are often confused. Comprehensionem: this word properly denotes the process of perception in the abstract, not the individual perception. The Greeks, however, themselves use καταληψις for καταληπτικη φαντασια very often. Quae manu prehenderentur: see II. 145. Nova enim dicebat: an admission not often made by Cic., who usually contends, with Antiochus, that Zeno merely renamed old doctrines (cf. 43). Sensum: so Stob., I. 41, 25 applies the term αισθησις to the φαντασια. Scientiam: the word επιστημη is used in two ways by the Stoics, (1) to denote a number of coordinated or systematised perceptions (καταληψεις or καταληπτικαι φαντασιαι) sometimes also called τεχνη (cf. Sext. Pyrrh. Hyp. III. 188 τεχνην δε ειναι συστημα εκ καταληψεων συγγεγυμνασμενων); (2) to denote a single perception, which use is copied by Cic. and may be seen in several passages quoted by Zeller 80. Ut convelli ratione non posset: here is a trace of later Stoicism. To Zeno all καταληπτικαι φαντασιαι were ασφαλεις, αμεταπτωτοι ‛υπο λογου. Later Stoics, however, allowed that some of them were not impervious to logical tests; see Sext. Adv. Math. VII. 253, qu. Zeller 88. Thus every καταληπτικη φαντασια, instead of carrying with it its own evidence, had to pass through the fire of sceptical criticism before it could be believed. This was, as Zeller remarks, equivalent to giving up all that was valuable in the Stoic theory. Inscientiam: ex qua exsisteret: I know nothing like this in the Stoic texts; αμαθια is very seldom talked of there. Opinio: δοξα, see Zeller and cf. Ac. II. 52, T.D. II. 52, IV. 15, 26.

§42. Inter scientiam: so Sextus Adv. Math. VII. 151 speaks of επιστημην και δοξαν και την εν μεθοπιαι τουτων καταληψιν. Soli: Halm, I know not why, suspects this and Christ gives solum ei. Non quod omnia: the meaning is that the reason must generalize on separate sensations and combine them before we can know thoroughly any one thing. This will appear if the whole sentence be read uno haustu; Zeller p. 78 seems to take the same view, but I have not come across anything exactly like this in the Greek. Quasi: this points out normam as a trans. of some Gk. word, κριτηριον perhaps, or γνωμων or κανων. Notiones rerum: Stoic εννοιαι; Zeller 81—84, R. and P. 367, 368. Quodque natura: the omission of eam is strange; Faber supplies it. Imprimerentur: the terms εναπεσφραγισμενη, εναπομεμαγμενη, εντετυπωμενη occur constantly, but generally in relation to φαντασιαι, not to εννοιαι. Non principia solum: there seems to be a ref. to those αρχαι της αποδειξεως of Arist. which, induced from experience and incapable of proof, are the bases of all proof. (See Grote's Essay on the Origin of Knowledge, first printed in Bain's Mental and Moral Science, now re-published in Grote's Aristotle.) Zeno's εννοιαι were all this and more. Reperiuntur: two things vex the edd. (1) the change from oratio obliqua to recta, which however has repeatedly taken place during Varro's exposition, and for which see M.D.F. I. 30, III. 49; (2) the phrase reperire viam, which seems to me sound enough. Dav., Halm give aperirentur. There is no MSS. variant. Aliena: cf. alienatos D.F. III. 18. A virtute sapientiaque removebat: cf. sapiens numquam fallitur in iudicando D.F. III. 59. The firma adsensia is opposed to imbecilla 41. For the adsensio of the sapiens see Zeller 87. More information on the subject-matter of this section will be found in my notes on the first part of the Lucullus. In his constitit: cf. II. 134.

§§43—end. Cicero's historical justification of the New Academy. Summary. Arcesilas' philosophy was due to no mere passion for victory in argument, but to the obscurity of phenomena, which had led the ancients to despair of knowledge (44). He even abandoned the one tenet held by Socrates to be certain; and maintained that since arguments of equal strength could be urged in favour of the truth or falsehood of phenomena, the proper course to take was to suspend judgment entirely (45). His views were really in harmony with those of Plato, and were carried on by Carneades (46).

§43. Breviter: MSS. et breviter; see 37. Tunc: rare before a consonant; see Munro on Lucr. I. 130. Verum esse [autem] arbitror: in deference to Halm I bracket autem, but I still think the MSS. reading defensible, if verum be taken as the neut. adj. and not as meaning but. Translate: "Yet I think the truth to be ... that it is to be thought," etc. The edd. seem to have thought that esse was needed to go with putandam. This is a total mistake; cf. ait ... putandam, without esse II. 15, aiebas removendum II. 74; a hundred other passages might be quoted from Cic.

§44. Non pertinacia aut studio vincendi: for these words see n. on II. 14. The sincerity of Arcesilas is defended also in II. 76. Obscuritate: a side-blow at declaratio 41. Confessionem ignorationis: see 16. Socrates was far from being a sceptic, as Cic. supposes; see note on II. 74. Et iam ante Socratem: MSS. veluti amantes Socratem; Democritus (460—357 B.C.) was really very little older than Socrates (468—399) who died nearly sixty years before him. Omnis paene veteres: the statement is audaciously inexact, and is criticised II. 14. None of these were sceptics; for Democritus see my note on II. 73, for Empedocles on II. 74, for Anaxagoras on II. 72. Nihil cognosci, nihil penipi, nihil sciri: the verbs are all equivalent; cf. D.F. III. 15 equidem soleo etiam quod uno Graeci ... idem pluribus verbis exponere. Angustos sensus: Cic. is thinking of the famous lines of Empedocles στεινοποι μεν γαρ παλαμαι κ.τ.λ. R. and P. 107. Brevia curricula vitae: cf. Empedocles' παυρον δε ζωης αβιου μερος. Is there an allusion in curricula to Lucretius' lampada vitai tradunt, etc.? In profundo: Dem. εν βυθω, cf. II. 32. The common trans. "well" is weak, "abyss" would suit better. Institutis: νομω of Democritus, see R. and P. 50. Goerenz's note here is an extraordinary display of ignorance. Deinceps omnia: παντα εφεξης there is no need to read denique for deinceps as Bentl., Halm. Circumfusa tenebris: an allusion to the σκοτιη γνωσις of Democr., see II. 73. Dixerunt: Halm brackets this because of dixerunt above, parts of the verb dicere are however often thus repeated by Cic.

§45. Ne illud quidem: cf. 16. Latere censebat Goer. omitted censebat though in most MSS. Orelli and Klotz followed as usual. For the sense II. 122. Cohibereque: Gk. επεχειν, which we shall have to explain in the Lucullus. Temeritatem ... turpius: for these expressions, see II. 66, note. Praecurrere: as was the case with the dogmatists. Paria momenta: this is undiluted scepticism, and excludes even the possibility of the probabile which Carneades put forward. For the doctrine cf. II. 124, for the expression Euseb. Praep. Evan. XIV. c. 4 (from Numenius) of Arcesilas, ειναι γαρ παντα ακαταληπτα και τους εις εκατερα λογους ισοκρατεις αλληλοις, Sextus Adv. Math. IX. 207 ισοσθενεις λογοι; in the latter writer the word ισοσθενεια very frequently occurs in the same sense, e g Pyrrhon. Hyp. I. 8 (add N.D. I. 10, rationis momenta)

§46. Platonem: to his works both dogmatists and sceptics appealed, Sextus Pyrrhon. Hyp. I. 221 τον Πλατωνα οιν ‛οι μεν δογματικον εφασαν ειναι, ‛οι δε απο ητικον, ‛οι δε κατα μεν τι απορητικον, κατα δε τι δογματικον. Stobaeus II. 6, 4 neatly slips out of the difficulty; Πλατων πολυφωνος ων, ουχ ‛ως τινες οιονται πολυδοξος. Exposuisti: Durand's necessary em., approved by Krische, Halm, etc. for MSS. exposui. Zenone: see Introd. p. 5.

NOTES ON THE FRAGMENTS.

BOOK I.

1. Mnesarchus: see II. 69, De Or. I. 45, and Dict. Biogr. 'Antipater'; cf. II. 143, De Off. III. 50. Evidently this fragment belongs to that historical justification of the New Academy with which I suppose Cicero to have concluded the first book.

2. The word concinere occurs D.F. IV. 60, N.D. I. 16, in both which places it is used of the Stoics, who are said re concinere, verbis discrepare with the other schools. This opinion of Antiochus Cic. had already mentioned 43, and probably repeated in this fragment. Krische remarks that Augustine, Cont. Acad. II. 14, 15, seems to have imitated that part of Cicero's exposition to which this fragment belongs. If so Cic. must have condemned the unwarrantable verbal innovations of Zeno in order to excuse the extreme scepticism of Arcesilas (Krische, p. 58).

BOOK II.

3. This fragm. clearly forms part of those anticipatory sceptical arguments which Cic. in the first edition had included in his answer to Hortensius, see Introd. p. 55. The argument probably ran thus: What seems so level as the sea? Yet it is easy to prove that it is really not level.

4. On this I have nothing to remark.

5. There is nothing distinctive about this which might enable us to determine its connection with the dialogue. Probably Zeno is the person who serius adamavit honores.

6. The changing aspects of the same thing are pointed to here as invalidating the evidence of the senses.

7. This passage has the same aim as the last and closely resembles Lucullus 105.

8. The fact that the eye and hand need such guides shows how untrustworthy the senses are. A similar argument occurs in Luc. 86. Perpendiculum is a plumb line, norma a mason's square, the word being probably a corruption of the Greek γνωμων (Curt. Grundz p. 169, ed. 3), regula, a rule.

9. The different colours which the same persons show in different conditions, when young and when old, when sick and when healthy, when sober and when drunken, are brought forward to prove how little of permanence there is even in the least fleeting of the objects of sense.

10. Urinari is to dive; for the derivation see Curt. Grundz p. 326. A diver would be in exactly the position of the fish noticed in Luc. 81, which are unable to see that which lies immediately above them and so illustrate the narrow limits of the power of vision.

11. Evidently an attempt to prove the sense of smell untrustworthy. Different people pass different judgments on one and the same odour. The student will observe that the above extracts formed part of an argument intended to show the deceptive character of the senses. To these should probably be added fragm. 32. Fr. 19 shows that the impossibility of distinguishing eggs one from another, which had been brought forward in the Catulus, was allowed to stand in the second edition, other difficulties of the kind, such as those connected with the bent oar, the pigeon's neck, the twins, the impressions of seals (Luc. 19, 54), would also appear in both editions. The result of these assaults on the senses must have been summed up in the phrase cuncta dubitanda esse which Augustine quotes from the Academica Posteriora (see fragm. 36).

BOOK III.

12. This forms part of Varro's answer to Cicero, which corresponded in substance to Lucullus' speech in the Academica Priora The drift of this extract was most likely this: just as there is a limit beyond which the battle against criminals cannot be maintained, so after a certain point we must cease to fight against perverse sceptics and let them take their own way. See another view in Krische, p. 62.

13. Krische believes that this fragment formed part of an attempt to show that the senses were trustworthy, in the course of which the clearness with which the fishes were seen leaping from the water was brought up as evidence. (In Luc. 81, on the other hand, Cic. drew an argument hostile to the senses from the consideration of the fish.) The explanation seems to me very improbable. The words bear such a striking resemblance to those in Luc. 125 (ut nos nunc simus ad Baulos Puteolosque videmus, sic innumerabilis paribus in locis esse isdem de rebus disputantis) that I am inclined to think that the reference in Nonius ought to be to Book IV. and not Book III., and that Cic., when he changed the scene from Bauli to the Lucrine lake, also changed Puteolosque into pisciculosque exultantes for the sufficient reason that Puteoli was not visible from Varro's villa on the Lucrine.

14. The passion for knowledge in the human heart was doubtless used by Varro as an argument in favour of assuming absolute knowledge to be attainable. The same line is taken in Luc. 31, D.F. III. 17, and elsewhere.

15. It is so much easier to find parallels to this in Cicero's speech than in that of Lucullus in the Academica Priora that I think the reference in Nonius must be wrong. The talk about freedom suits a sceptic better than a dogmatist (see Luc. 105, 120, and Cic.'s words in 8 of the same). If my conjecture is right this fragment belongs to Book IV. Krische gives a different opinion, but very hesitatingly, p. 63.

16. This may well have formed part of Varro's explanation of the καταληψις, temeritas being as much deprecated by the Antiocheans and Stoics as by the Academics cf. I. 42.

17. I conjecture malleo (a hammer) for the corrupt malcho, and think that in the second ed. some comparison from building operations to illustrate the fixity of knowledge gained through the καταληψεις was added to a passage which would correspond in substance with 27 of the Lucullus. I note in Vitruvius, quoted by Forc. s.v. malleolus, a similar expression (naves malleolis confixae) and in Pliny Nat. Hist. XXXIV. 14 navis fixa malleo. Adfixa therefore in this passage must have agreed with some lost noun either in the neut. plur. or fem. sing.

18. This and fragm. 19 evidently hang very closely together. As Krische notes, the Stoic εναργεια had evidently been translated earlier in the book by perspicuitas as in Luc. 17.

19. See on Luc. 57.

BOOK IV.

Further information on all these passages will be found in my notes on the parallel passages of the Lucullus.

21. Viam evidently a mistake for the umbram of Luc. 70.

23. The best MS. of Nonius points to flavum for ravum (Luc. 105). Most likely an alteration was made in the second edition, as Krische supposes, p. 64.

28. Corpusculis: Luc. 121 has corporibus. Krische's opinion that this latter word was in the second edition changed into the former may be supported from I. 6, which he does not notice. The conj. is confirmed by Aug. Contr. Ac. III. 23.

29. Magnis obscurata: in Luc. 122 it is crassis occultata, so that we have another alteration, see Krische, p. 64.

30. Only slight differences appear in the MSS. of the Luc. 123, viz. contraria, for in c., ad vestigia for contra v.

31. Luc. 137 has dixi for dictus. As Cic. does not often leave out est with the passive verb, Nonius has probably quoted wrongly. It will be noted that the fragments of Book III. correspond to the first half of the Luc., those of Book IV. to the second half. Cic. therefore divided the Luc. into two portions at or about 63.

UNCERTAIN BOOKS.

32. I have already said that this most likely belonged to the preliminary assault on the senses made by Cic. in the second book.

33. In the Introd. p. 55 I have given my opinion that the substance of Catulus' speech which unfolded the doctrine of the probabile was incorporated with Cicero's speech in the second book of this edition. To that part this fragment must probably be referred.

34. This important fragment clearly belongs to Book II., and is a jocular application of the Carneadean probabile, as may be seen from the words probabiliter posse confici.

35. Krische assigns this to the end of Varro's speech in the third Book. With this opinion I find it quite impossible to agree. A passage in the Lucullus (60) proves to demonstration that in the first edition this allusion to the esoteric teaching of the Academy could only have occurred either in the speech of Catulus or in that of Cicero. As no reason whatever appears to account for its transference to Varro I prefer to regard it as belonging to Cic.'s exposition of the positive side of Academic doctrine in the second book. Cic. repeatedly insists that the Academic school must not be supposed to have no truths to maintain, see Luc. 119, also 66 and N.D. I. 12. Also Aug. Contra. Ac. II. 29.

36. It is difficult to see where this passage could have been included if not in that prooemium to the third book which is mentioned Ad. Att. XVI. 6, 4. I may here add that Krische seems to me wrong in holding that the whole four books formed one discussion, finished within the limits of a single day. Why interrupt the discussion by the insertion of a prologue of so general a nature as to be taken from a stock which Cic. kept on hand ready made? (Cf. Ad Att. as above.)

Besides the actual fragments of the second edition, many indications of its contents are preserved in the work of Augustine entitled Contra Academicos, which, though written in support of dogmatic opinions, imitated throughout the second edition of the Academica of Cic. No writings of the Classical period had so great an influence on the culture and opinions of Augustine as the Academica and the lost Hortensius. I give, partly from Krische, the scattered indications of the contents of the former which are to be gathered from the bishop's works. In Aug. Contr. Ac. II. 14, 15, we have what appears to be a summary of the lost part of Book I. to the following effect. The New Academy must not be regarded as having revolted against the Old, all that it did was to discuss that new doctrine of καταληψις advanced by Zeno. The doctrine of ακαταληψια though present to the minds of the ancients had never taken distinct shape, because it had met with no opposition. The Old Academy was rather enriched than attacked by the New. Antiochus, in adopting Stoicism under the name of the Old Academy, made it appear that there was a strife between it and the New. With Antiochus the historical exposition of Cic. must have ended. From this portion of the first book, Aug. derived his opinion (Contra. Ac. II. 1) that New Academicism was excusable from the necessities of the age in which it appeared. Indications of Book II. in Aug. are scarce, but to it I refer Contra. Ac. I. 7 placuit Ciceroni nostro beatum esse qui verum investigat etiam si ad eius inventionem non valeat pervenire, also ibid. III. 10 illis (Academicis) placuit esse posse hominem sapientem, et tamen in hominem scientiam cadere non posse. These I refer to Cicero's development of the probabile in Book II., although I ought to say that Krische, p. 65, maintains that the substance of Catulus' exposition in the Ac. Priora transferred to Book IV. of the Ac. Posteriora. As this would leave very meagre material for Book II., nothing indeed excepting the provisional proof of the deceptiveness of the senses, I cannot accede to his arrangement; mine, I may remark, involves a much smaller departure from the first edition. Allusions in Aug. to the attack on the senses by Cic. in Book II. are difficult to fix, as they apply equally well to the later attack in Book IV. As to Books III. and IV., I do not think it necessary here to prove from Aug. the points of agreement between them and the Lucullus, which will find a better place in my notes on the latter, but merely give the divergences which appear from other sources. These are the translation of σοφισματα by cavillationes in Luc. 75 (Seneca Ep. III.), and the insertion in 118 of essentia as a translation of ουσια.