Perhaps never has a philosophical work of Cicero received such lavish treatment within such a short time from such distinguished scholars as the Academici libri have in the past two years, beginning with Tobias Reinhardt’s 2023 OCT and commentary (for the latter, cf. BMCR 2023.11.21) and continuing with this volume. This review will focus on the text, for which Hunt bears chief responsibility, in comparison with that of Reinhardt, which has not been reviewed in this journal.
The edition is preceded by an elaborate preface, including a section on the history of the text and principles of the edition by Hunt (pp. cccxxx-ccclxix). Here, relying on the findings of Hunt 1998, he arrives at a stemma (p. cccxlvii) with two hyparchetypes, γ and P, followed in essentials also by Reinhardt (I refer to witnesses with the Budé sigla, since that is the work under review). It is shown that of the items in Cicero’s philosophical corpus, Ac. 1 is preserved in mss. fewest in number and most recent in date, a fact that would argue for a rather skeptical attitude toward the transmitted text.
1 atque illum complexi, ut mos amicorum est, satis eum/enim longo interuallo ad suam uillam reduximus.
Cicero, Atticus, and Varro meet and greet each other halfway between the estates of Cicero and Varro. The split between γ (enim) and P (eum) creates uncertainty in this bifid tradition, Reinhardt opting for eum, Hunt for enim. Then Hunt must, however, take satis enim longo intervallo as a separate clause lacking a finite verb and posit a lacuna, as Plasberg had when he proposed adding <aduenerat> after interuallo. But this creates an improbable relation of clauses; cf. Reinhardt 2023: 93: “a remark that they had not seen each other for a long time would… jar, in that it would amount to an apology for an… overcordial embrace.” There are various ways of referring to Varro in this passage, but this is not a problem, since each reference is appropriate (contrast Hunt, 53): first ipsum (“the man himself,” i.e., the man they were looking for), then illum complexi, still emphatic but slightly less so, and finally the unemphatic eum as object of the routine action
2 hic pauca primo atque ea percunctantibus nobis ecquid forte Roma novi, Atticus… inquit…
atque ea is the transmitted text, defended by Reinhardt 2023: 94 on the ground that it is a formula for adding an emphatic and/or particularizing detail (cf. Kühner and Stegmann 1:619). In support he cites Dom. 98: suscipere tantos animi dolores atque ea quae capta urbe accidunt victis. But there the neuter plural functions as the antecedent of quae, whereas in our text ea is followed by an ablative absolute. To clinch the argument, we need evidence that atque and a form of is can be followed by an adverbial expression; cf. §25 atque id in multis; Sen. 74: sensus moriendi aliquis esse potest, isque ad exiguum tempus. No need, then, to delete ea with some recentiores, followed by Hunt. However, Reid’s insertion of tum before Atticus, accepted by Reinhardt but not Hunt, is unnecessary. Cicero’s elaborate sentences are not infrequently broken up by modern editors, contrary to the author’s preferences.
2 habeo magnum opus in manibus, quae iam primum.
Varro is the speaker. Hunt adopts Christ’s <id>que for quae or que transmitted here, whereas Reinhardt retains the text printed above, citing parallels adduced ad loc. by Plasberg 1908. Three of these involve ellipsis of the verb in the relative clause. The real problem, however, is the shift from singular to plural from antecedent to relative pronoun, for which Plasberg cites Luc. 126, Tusc. 1.6, and De orat. 1.226, but in these instances the pronoun and antecedent are at some distance, not so close as here, and we have the additional problem, pointed out by Hunt, that manibus could (theoretically) be taken as the antecedent of quae. Once quae established itself as the predominant reading (perhaps in view of quaedam and another quae in the vicinity), a scribe would be tempted to delete the puzzling id.
5 (a) uides autem eadem ipse: didicisti enim non posse nos Amafinii aut Rabirii similes esse…
The above is the transmitted text and is printed by Hunt. Davies, however, proposed the transposition of enim as follows:
(b) uides autem – eadem enim ipse didicisti – non posse nos Amafinii aut Rabiri similes esse…
Both Reid and Reinhardt have adopted this text, rightly, I think (I assume that Reinhardt’s omission in his OCT of the first dash is merely a typo, since the above is the text he defends in his commentary); Hunt does not cite Davies’ conjecture in his apparatus or discuss it in the endnotes. The problem is the two verbs uides and didicisti and what they might plausibly refer to and the relation of the clause connected with enim.
In (a), with eadem as object of uides, Cicero “sees” the same point, i.e., the dilemma Varro has just propounded that writing on philosophy in Latin would be useless for the learned and the unlearned alike because he has “learned” that he and Varro (nos) cannot be like the named Epicurean authors (criticized in the sequel for lack of argumentative and stylistic rigor).
In (b), the inability of Varro and Cicero to be like the Epicurean authors is what Cicero “sees,” and the reason is that he and Varro have “learned” the same things (i.e., at the feet of Antiochus of Ascalon). Surely this gives the more plausible sense to both verbs. The “learning” would be philosophical education; cf. §6, where nostra tu physica nosti is the premiss to the argument that physics is a difficult subject. (Reinhardt 2023: 99, points out that at Tusc. 2.7 Cicero says that he declined to read Latin authors of philosophical tracts, which would further undermine any claim that he has “learned” of the unbridgeable dissimilarity of the Epicurean authors; the argument put in Varro’s mouth here shows, however, at least a superficial acquaintance with them.) enim would have explanatory power in that Varro’s and Cicero’s shared philosophical education set them apart from the likes of Amafinius and Rabirius.
Under (a), on the other hand, it is obscure how Cicero is supposed to have “learned” that he and Varro cannot be similar to the named Epicureans and how that explains (enim) his ability to “see” Varro’s dilemma. The transmitted text may have its origin in a saltation error, a scribe’s eye having skipped from the m of eadem to that of enim, with the missing word added to the margin or interlinearly and falsely inserted in subsequent copying.
8 in laudationibus, in his ipsis antiquitatum pro<h>oemiis philosophiae scribere uoluimus.
The Budé retains the transmitted philosophiae with the rendering “j’ai voulu écrire pour la philosophie.” But one expects scribo with dative of person (cf. OLD 13a, 14b). On the other hand, Reinhardt adopts Plasberg’s conjecture philosophiae <more>. But with more one expects genitive of the person, as Hunt, 77n60, points out. Taking up Casaubon’s suggestion, cited by Hunt, one might rather expect a substantive, e.g., initia, to have dropped out prior to philosophiae (elementa appears to be excluded because of the fuss Varro makes in using the term at §26). This would accord with Varro’s generally modest tone in this work. For pro<h>oemiis, see 75–76n58 (Revello).
9 philosophiam… inchoasti, ad impellendum satis, ad edocendum/educandum parum.
“Cicero” addresses Varro. Hunt reads edocendum, Reinhardt educandum; neither indicates a variant. I have checked the Madrid ms. online (Bibl. Nac. 9116, fol. 257r) and can confirm Hunt’s reading.
11 ego autem/Varro… animo haec inclusa habebam… et renouabam cum licebat legendo.
“Cicero” on his relation to philosophical studies. Again the tradition is split between P (autem) and γ (uarro). Reinhardt follows Plasberg in adopting both readings. But Hunt rightly objects that Cicero’s use of the vocative is limited to instances where there is a change of speaker (82n73).
17 Nam cum Speusippum sororis filium Plato philosophiae quasi heredem reliquisset, duo autem praestantissimo studio atque doctrina Xenocratem Calchedonium et Aristotelem Stagiritam <quasi procuratores>…
Reinhardt’s insertion of quasi procuratores, though bold, is attractive; certainly the transmitted text strikes one as incomplete. Cicero, who tends to strive for fullness and balance, might be expected to explain the role of Xenocrates and Aristotle just as he has that of Speusippus. (The Budé does not cite the conjecture or discuss it in the endnotes.)
21 hominem (enim) esse censebant quasi partem quandam ciuitatis et uniuersi generis humani.
enim occurs in the γ hyparchetype but is absent from P. A further complication is that the two branches of γ diverge, φ reading hominem enim esse, whereas ε presents esse before ciuitatis. I agree with Plasberg and Reinhardt that enim yields good sense; as Plasberg 1908: 41 writes, “sed enim peraptum est, quoniam illa quae beatae uitae adiuncta sunt, i.e. patria parentes liberi amici alia, humana societate continentur.” If that is so, we should assume that enim dropped out of P, perhaps by saltation after hominem, and that the ε branch of γ was independently corrupted by loss of esse (perhaps by homoearcton after enim), which was then added to the margin or interlinearly and falsely reinserted. A similar case is §25, where both Reinhardt and Hunt prefer nobis magis (φ), comparing Fin. 3.15, to P’s magis (corrected from maius) nobis and ε’s nobis maius. Denying enim to the archetype, the Budé prints the text without a conjunction but does not argue for asyndeton; rather, Hunt and Malaspina weigh other connectors as possible replacements (pp.100–1), but none of these is as convincing as enim.
29 quam vim animum esse dicunt mundi… quam interdum eandem necessitatem appellant, quia nihil aliter possit atque ab ea constitutum sit, inter quasi fatalem et immutabilem continuationem ordinis sempiterni; non numquam quidem eandem fortunam…
An elaborate description of the natural force the Stoics identified as god. The difficulty begins with inter. Hunt says it is rare in the sense “in” (“dans”: 114n164) but offers no parallel; indeed, one is hard-pressed to find a Ciceronian one (cf. TLL 7.1, 2144.82 ff.). Plasberg conjectured inter<dum> and set this in his text, marking a following lacuna and suggesting in his apparatus seriem causarum (a phrase taken from Div. 1.125); Reinhardt sets both interdum and seriem causarum in his text. But if this is to be on the same level as eandem esse mentem above and non numquam… eandem fortunam below, one would expect inter<dum eandem seriem causarum>. But even this is unsatisfactory, since such a series of causes would hardly be on the same level as necessitas and fortuna (similarly Hunt, l.c.); it might more plausibly be a part of the description of the former; cf. N.D. 1.55: hinc vobis extitit… illa fatalis necessitas, quam εἱμαρμένη dicitis, ut quicquid accidat id ex aeterna veritate causarumque continuatione fluxisse dicatis. I would therefore tentatively suggest substituting propter for inter (and deleting the comma after sit): if the beginning of the word were obscured or otherwise difficult to read, a scribe could easily fall into inter in this context.
33 Haec erat illis pri<ma philosophiae for>ma, a Platone tradita, cuius quas acceperim disputationes, si uultis, exponam.
Above is the Budé text, incorporating Malaspina’s excellent supplement of P’s haec erat illis prima (Reinhardt follows Madvig’s haec forma erat illis prima based on γ’s haec prima erat illis prima). disputationes is difficult, especially with objective genitive (though it is paralleled at Agr. 2.40, cited by Hunt, 130). Perhaps it can stand in the general sense “discussions.” It is hard to see how Davies’ immutationes, though attractive in sense (and set in the text by Reinhardt), could have been so corrupted; it may be a correction of the author.
36–37 Quae autem secundum naturam essent ea sumenda et quadam aestimatione dignanda docebant, contraque contraria;… Sed quae essent sumenda, ex iis alia pluris esse aestimanda, alia minoris <contraque contraria>. Quae pluris ea praeposita appellabat, reiecta autem quae minoris.
From Varro’s summary of Stoic ethics. contraque contraria is Reinhardt’s insertion to complete the sense, since both positive and negative values are under discussion; he suggests that the words might have dropped out because they were “deemed to be an erroneous repetition from above” (2023: 224); alternatively, there may have been a saltation error from -ris to -ria. Hunt cites the conjecture in his apparatus but does not discuss it in the endnotes. If one were to defend the transmitted text, it would be on the ground of Cicero’s haste/carelessness in composition.
41 Quod autem erat sensu comprensum… si ita erat comprensum ut conuelli ratione non posset scientiam, sin aliter inscientiam nominabat, ex qua existeret etiam opinio, quae esset imbecilla et cum falso incognitoque communis.
From the exposition of Stoic epistemology. In his apparatus Plasberg suggested supplying adsensio after imbecilla, and this is set in the text by Reinhardt. Though there would be no philosophical objection to the supplement, as Reinhardt shows (2023: 243), imbecilla and communis can stand as parallel predicates (and imbecilla, a double trochee, is the better clausula); no change is indicated.
43 Verum esse autem arbitror, ut Antiocho nostro familiari placebat, correctionem ueteris Academiae potius quam aliquam nouam disciplinam putandam.
At issue is the status of Stoicism. For transmitted uerum Plasberg and Reinhardt adopt horum, which is perhaps no improvement. The subject of putandam is not explicitly stated; rather, Stoicorum rationem must be understood from the previous sentence. However, that fault, if such it be, is compounded if one adds horum as a modifier of the understood subject. The transmitted text makes good sense (“But I think it is true, as our friend Antiochus held, that it must be considered a correction of the Old Academy, not some new doctrine”); similarly Hunt and Malaspina, 154–55.
45 Itaque Arcesilas negabat esse quicquam quod sciri posset, ne illud quidem ipsum quod Socrates sibi reliquisset, ut nihil scire se sciret.
Reinhardt deletes as a gloss ut nihil scire se sciret. Though these words are omitted in P and in one branch on the other side of the stemma, μ, they must be assumed as the reading of the other hyparchetype, γ, so we have another faultline running between the two branches. I would keep the words in the text (as does Hunt) for these reasons: (a) Cicero generally tends to fullness of expression; (b) he might not have felt able to trust his Roman readers to supply the point on their own; (c) one should be loath to delete the cretic plus trochee scire se sciret.
47 Hanc Academiam nouam appellant, quae mihi uetus uidetur… sed tamen illa quam exposui uetus, haec noua nominetur;…
exposui is the transmitted text, but is difficult since it was manifestly Varro who expounded the doctrines of the Old Academy; hence Durand’s exposui<sti>, which is adopted by Reinhardt. Hunt retains the transmitted text (but marks it as corrupt) as an author’s error. Possibly so, but it could be a simple saltation error (and the manuscript tradition is generally not strong; see above); I would be inclined to emend.
The above is an assessment of major divergences in the texts of two able editors, one a bit bolder, the other more cautious. This framework precludes, however, doing justice to Hunt’s many valuable notes on niceties of Latinitas, let alone Lévy’s rich notes on the content. The Academicus I is now clearly on a better footing, but work remains to be done.
Other Editions Cited
Kühner R. and C. Stegmann. 1966. Ausführliche Grammatik der lateinischen Sprache. Vol. 2: Satzlehre. 4th ed., ed. A. Thierfelder. 2 vols. Darmstadt.
Plasberg, O. 1908. M. Tulli Ciceronis Paradoxa Stoicorum, Academicorum reliquiae cum Lucullo etc. Leipzig.
Plasberg, O. 1922. M. Tulli Ciceronis Academicorum reliquiae cum Lucullo. Leipzig.
Reid, J.S. 1885. M. Tulli Ciceronis Academica. London.
Reinhardt, T. 2023. M. Tulli Ciceronis Academicus primus, fragmenta et testimonia Academicorum librorum, Lucullus. Oxford.
Other References
Hunt, T.J. 1998. A Textual History of Cicero’s Academici libri. Leiden.
Reinhardt, T. 2023. Cicero’s Academici libri & Lucullus: A Commentary with Introduction and Translations. Oxford.