Antonio Cassarino ’ s Latin translation of Plutarch ’ s Bruta animalia ratione uti

The present paper aims at giving a text edition of Antonio Cassarino ’ s humanist Latin translation of Plutarch ’ s dialogue Bruta animalia ratione uti . This is the earliest of three translations made of this dialogue in the course of the 15th century. The text itself is extant in three different manuscripts, one of which is a codex of the Vatican Library (Vat. lat. 3349), compiled after Cassarino ’ s death by Panormita. A comparison of the text variants has led to several results. First, some of the errors shared by all three manuscripts show that they go back to a common archetype already at some distance from the translator ’ s original copy. Second, Panormita relied heavily on a codex of the Biblioteca della Società Siciliana per la Storia Patria in Palermo (MS Lodi XII E 13) in preparing his own version. Third, the Vatican codex is far from being the best representative of Cassarino ’ s original translation. Though Panormita corrected several of the common inherited errors, he made changes to the text without consulting the Greek. In almost every instance, it is a codex of the Biblioteca Casanatense of Rome (Bibl. Casan. 665 C II 8) which gives the best reading, providing the clue for a successful reconstruction of the text. An attempt will be made to trace the version contained in this codex back to a certain person named Balbi, referred to in the dedicatory letter as being a learned expert of both languages, Greek and Latin. Along with the establishment of the text, it will also be possible to de ﬁ ne the original Greek source codex Cassarino used for his translation (Vat. Pal. gr. 170).

The original Greek source which served as a basis for Giovanni's translation, as opposed to Raffaele's revised version, must have been close to the text of codex Z, 10 belonging to family Θ of Plutarch 's Moralia. 11 This conclusion is warranted by a number of isolated readings shared by the original Greek and the translation. 12Raffaele, on the other hand, used for his revised version texts which rather correspond to the codices of family Π, named after Planudes, 13 and the closely related FJ branch.
In the use of vocabulary Giovanni shows visible signs of dependence from Lampugnino Birago's earlier translation.The great number of correspondences between word choices cannot be ascribed to mere coincidence. 14Therefore, we must believe that Giovanni exploited Birago's translation in order to avail himself of a set of lexical meanings without, however, directly imitating his sentence constructions.The main challenge remained for him to build the sentences on his own and reformulate the text in accordance with a more elegant Latin style.
Lampugnino Birago's translation of the Bruta animalia can be assigned to the years between 1465 and 1470. 15Dedicated to Pope Paul II, it is extant in two manuscripts, one of which is preserved in the Vatican Library, 16 and the other in the Biblioteca Angelica of Rome. 17As inner evidence shows the translation was prepared on the basis of a Greek text characterized by the individual readings of codex A, 18 belonging to family Π.Among the number of copies known to us today codices βγδ 19 are the most likely candidates for having been the original source of Birago's Latin translation. 20he most characteristic feature of Birago's translation is his literal mode of rendering originating in medieval practice as well as a tendency to always keep the same word in Latin wherever a given word occurs in the Greek.What appears to lie behind this meticulous approach is the translator's adherence to lexical meanings furnished by contemporary dictionaries.As regards word usage one can distinguish in Birago's translation two different layers, both of which reflect the use of a specific class of dictionaries.In a large number of occurrences, the meanings of words are unmistakably taken from Pseudo-Cyril's Byzantine dictionary which was known to humanist circles through its arrival to Italy in the 1430s.Another group which was evidently used by Birago is the cluster of dictionaries, at his time only accessible in manuscripts, which eventually led to the creation of Giovanni Crastone's printed Greek-Latin lexicon 10 Venice, Biblioteca Nazionale Marciana, gr.Z. 511.11   Family Θ of Plutarch's Moralia comprises codices ZυB, to which codex b can also be added as a possible source.

13
A family which contains codices αAEG.
14 For details, see GAÁL (n.12) 227-234.15   Lampugnino Birago was born towards the end of the 14th century and lived until 1472.He spent the first half of his life in Milano before he went over to Rome to live the rest of his life in papal service.16   Vatican City, Biblioteca Apostolica Vaticana, Vat.lat.1887, 80 v -92 v .17 Rome, Biblioteca Angelica, Ms 1354, 71 v -82 r .The manuscript was first mentioned by BECCHI (n. 9) 156, n. 5.
As a matter of fact, Birago's rendering of the dialogue is only one in the series of five works chosen by him to be translated from Plutarch's Moralia and dedicated to the pope. 22It has been noted several times by scholars that Birago's choice of the works to be translated must have been influenced by Antonio Cassarino. 23There is not a single title in Birago's translation which was not included in Cassarino's former translation of nine Plutarchean works. 24We even find an almost complete agreement in the order of the works included.It is also remarkable that the works which were left untranslated by Birago had all been taken up for translation by other humanists during the period between Cassarino and Birago, a fact that would account for their omission by Birago. 25It is really difficult to believe that the idea to translate a work like the Bruta animalia would have occurred to Birago independently of Cassarino.The first impulse to translate this and other works of Plutarch must have been given by Cassarino whose translation of the Bruta animalia does lie at the root of all subsequent translations made of this dialogue in the course of the 15th century.

THE LATIN CODICES OF CASSARINO'S TRANSLATION
As part of a larger plan to translate the whole of Plutarch's Moralia, Antonio Cassarino brought his translation of nine Plutarchean works to completion in the years between 1440 and 1445. 26e had spent a number of years prior to this period in Constantinople where he was able to improve his knowledge of Greek.Upon returning to Italy he found employment as a schoolmaster in Genoa, receiving a public salary.But he could not make a secure living as his salary was reduced time and again.Still, these years saw Cassarino in his most productive stage of life when he eagerly devoted himself to translation.As years passed things became worse and Cassarino would have desired to change his place of residence, seeking help from his friend Antonio Panormita in accomplishing this end.Ill fate plotted against him, since in a state of commotion when some enraged citizens broke into his house and forced him to escape through the window he fell to the ground and was killed outright.

25
De capienda ex inimicis utilitate and Regum et imperatorum apophthegmata were translated by Janus Pannonius, Septem sapientium convivium by Giovanni Aurispa and Niccolò Perotti.De cupiditate divitiarum is a different case for it only sporadically appears in manuscripts and must therefore have remained unknown to Birago.After Cassarino's death the translations of Plutarch's works prepared until then were taken care of by his fellow Sicilian Panormita.What is now left behind of Cassarino's literary activity is for the most partthough not exclusivelycontained in two codices of the Vatican library, 27 compiled and arranged under Panormita's editorship.One of these codices lists eight of the nine Plutarchean works translated by Cassarino.The list of the works is as follows: V 5 Vatican City, Biblioteca Apostolica Vaticana, Vat.lat.3349. 281 r -12 r De laude ipsius, 12 v -44 v Quomodo adulator ab amico internoscatur, 44 v -53 v Bruta animalia ratione uti, 53 v -62 v De capienda ex inimicis utilitate, 62 v -78 r De cohibenda ira, 78 r -101 r Septem sapientium convivium, 101 r -142 v Regum et imperatorum apophthegmata, 147 r -184 v Apophthegmata Laconica.
A manuscript in the Biblioteca della Società Siciliana per la Storia Patria gives the same order of works with a slight alteration, even though some of the texts in this collection have apparently suffered loss in the process of transmission.The codex has the following titles: L 5 Palermo, Biblioteca della Società Siciliana per la Storia Patria, MS Lodi XII E 13. 29 2 r -8 r De laude ipsius, 8 v -9 v Quomodo adulator ab amico internoscatur (translation missing), 21 r -31 v Septem sapientium convivium (dedication missing), 32 r -36 v Bruta animalia ratione uti, 36 v -40 r De capienda ex inimicis utilitate, 40 v -48 r De cohibenda ira, 48 v -69 v Apophthegmata Laconica (end of translation missing).
In another manuscript which is kept in the Biblioteca Casanatense of Rome only two of Cassarino's translations have survived.These are part of a larger codex containing works of varied topics and origins, some of them betraying close links with Genoa.The relevant translations are: C 5 Rome, Biblioteca Casanatense 665 (C II 8). 3056 r -62 v Bruta animalia ratione uti, 64 r -94 r Regum et imperatorum apophthegmata.
Finally, mention must be made of two codices which are less important for our purposes.These fill up a gap by adding the one missing work De cupiditate divitiarum to the series of translations known from the other sources. 31Since, however, neither of these manuscripts contains the dialogue Bruta animalia they will be left out of the discussion here.The present analysis of the three aforementioned manuscripts will strictly focus on the text of the Bruta animalia.Cassarino's translation of this dialogue is dedicated to his disciple Giacomo Curlo who in the last lines of the dedicatory letter is advised to be careful with passing on the translation to anyone before consulting Balbi, the greatest expert in both languages.The latter should be able to correct whatever imperfections there are in the phrasing or explain unclear passages. 32In a rather formulary manner it is also stated that for lack of time the translation had been completed during a short night's work. 33eyond the possible imperfections resulting from the translator's haste in doing the translation or making occasional mistakes, there is a chance that a number of errors due to scribal copying or even traces of emendation intended for obtaining a better sense will be found in some or all of the manuscripts.The task lying before us is, therefore, to specify the differences between the codices by making a record of the variant readings and reach as close to the text of Cassarino's original translation as possible.
Difficulties, however, arise at the very first steps.In contrast to what might be expected, the codices are not exempt from a number of common errors which must necessarily be taken as proof of their descending from a common archetype with an already partially corrupt text.It is unimaginable that this common archetype should have originated with Cassarino, otherwise he would have corrected the errors himself.There must have been at least one phase in the line of transmission which separated Cassarino's original copy [x] from the common archetype [α] of the three manuscripts: Judging by the number of errors, however, it is even more likely that there existed more than one intermediate phases leading up to the common archetype [α]: 34 The first group of errors common to each of the three codices provides a series of banal mistakes deriving from scribes' misreading or miswriting words in the course of reproduction.They 32 Tu cave prius cuiquam hanc communices quam Balbum nostrum utriusque linguae doctissimum consulueris.Eius enim auxilio poteris non modo si qua minus commode dicta erunt emendare, sed si qua etiam obscuriora videbuntur intellegere.

33
Sed cum hoc tempore aliquid ad te dare statuissem, nec otii mihi satis (nam occupationes meas nosti) ad id quod volebam suppeteret, disputationem hanc brevissima lucubratiuncula e Plutarcho tibi transtuli… For the whole text of the dedication with a treatment of some problems like Balbi's identity, see RESTA (n.23) 233-234; cf. also INDELLI: Traduzioni (n. 3) 232.34   Some orthographic features shared by the three manuscripts may belong to this category as well.It is not easy to tell whether they come from Cassarino or reflect the conventional use of the scribes.Two of the most spectacular examples are the typical Italian word forms aiuta < adiuta and olfatus < olfactus.Furthermore, there is a frequent occurrence of itacism, dropping and metathesis of aspiration, simplification of double consonants, duplication of single consonants, irregularities of vowel and consonant assimilations etc. are generally easy to restore by having recourse to the Greek.The instances are given in the order of their occurrence in the text: 986C restitutus VLC < restitutum; 987E atris VLC 35 < acris; 988A malitia VLC < militia; 988B formata VLC < formatam; 988C vitae VLC < vitat; 988D nobis VLC < vobis; 989F tapetiis VLC < tapetis; 990F coire VLC < coiere; 990F quarundam VLC 36 < quorundam.
In a second group we find Greek proper names the correct spelling of which must have escaped the copyist.Some of the instances attest to several stages of corruption and have variant forms in the codices: 987F Kρoμμυωνίαν Cronimiam VLC < Crommiam < Crommyoniam; 988B Kαρίνας Caunas V Cairnas LC < Carinas; 992D Aὐτoλύκῳ Anthiloco VL Ancolyco C < Autolyco.
A third group is made up of a certain number of graver textual errors.These in the first place include two passages where parts of a sentence appear to have fallen out of the text.The first is 988E πῶς γὰρ oὐκ ἄτoπoν αἰτιᾶσθαι μὲν ὑμᾶς τὴν φύσιν, ὅτι μὴ κέντρα πρoσέφυσε τoῖς σώμασι μηδὲ ἀμυντηρίoυς ὀδόντας μηδὲ ἀγκύλoυς ὄνυχας, αὐτoὺς δὲ τῆς ψυχῆς τὸ σύμφυτoν ἀφαιρεῖν ὅπλoν καὶ κoλoύειν; Vel quonam modo absurdum non est naturam vos accusare quod neque aculeis nec dentibus aut unguibus corpus armaverit, cum naturale, ut dixerim, telum eripiatis et †quodammodo †? VLC.The word quodammodo put between cruces in the translation might have entered the text through scribal error under the influence of quonam modo at the beginning of the sentence.The presence of the conjunctive particle et is indicative of the loss of certain parts of the sentence.What must be missing from the translation is the Greek κoλoύειν and τῆς ψυχῆς, therefore we might tentatively supplement <decurtetis animo>. 37he second passage to be cited in this group is 990A ἡ δ' ὄσφρησις ἡμῶν πρὸ τῶν χυμῶν γνώμων oὖσα τῆς δυνάμεως ἑκάστoυ πoλὺ τῶν βασιλικῶν πρoγευστῶν σκεπτότερoν διαισθανoμένη, τὸ μὲν oἰκεῖoν εἴσω παρίησι, τὸ δὲ ἀλλότριoν ἀπελαύνει… Noster siquidem olfactus uniuscuiusque naturam et multo acrius quam regum praegustatores, quod naturae suae conveniens est admittit, quod alienum atque incongruens respuit… VLC.Here in the first clause the translation of the words πρὸ τῶν χυμῶν γνώμων oὖσα is missing and another Greek participle διαισθανoμένη is also left without translation.However, the causal adverb siquidem introducing the sentence must have originally been used in conjunction with a finite verb to make the sense complete.We might therefore propose an insertion like Noster siquidem olfactus uniuscuiusque naturam <cognoscit ante sapores>… Things are even more complicated in three further passages where attempts at emendation of the text are to be seen in one or two of the variants.There seems to be some confusion in the tradition of a certain line in 985E ῞Oτι νὴ Δία καλὴν ἄν μoι δoκῶ γενέσθαι φιλoτιμίαν πρὸς τoὺς ῞Eλληνας, εἰ χάριτι σῇ λαβὼν τoύτoυς αὖθις ἀνθρώπoυς ἑταίρoυς ἀνασώσαιμι… Quia honestum, mediusfidius, studium erga meos declaraturum me arbitror, si hos a te dono accipiens rursus homines, socios salvus compotes fecero… LC 38 rursus homines, salvos compotes fecero… V.The problem may have arisen from the perplexing apposition of three accusatives in the Greekτoύτoυς, ἀνθρώπoυς, ἑταίρoυςnot easy to render. 39 striving to obtain by Circe's favour to set them free after they are changed back into human form as his comrades.Of the Latin equivalents V preferred dropping socios and replaced the grammatically incorrect form salvus by salvos.Still, it remains unclear what role compotes has to play, a word which appears out of place here. 40n the second and third passages only a recourse to textual conjecture can save the situation.The Greek original may be of help to us.We read in 989C τὰ δὲ θηρία παντάπασιν ἀβάτoυς καὶ ἀνεπιμίκτoυς ἔχoντα τoῖς ἐπεισάκτoις πάθεσι τὰς ψυχὰς… Animalia vero, cum sint omnino a defectionibus libera… V Animalia vero, cum sint ab externis omnino a deffectionibus libera… L 41 Animalia vero, cum sint ab externis omnino a deffensoribus libera… C. 42 It would be unreasonable to follow V and get rid of ab externis which has its exact counterpart in the Greek verbal adjective ἐπεισάκτoις.The latter must be taken together with the dative plural πάθεσι in the sentence.Accordingly, πάθεσι is the only possible member to be identified with deffensionibus or any other similar form.This will then lead us to suppose that a corruption like a deffensionibus < affectionibus might have taken place.The whole clause would read: Animalia vero cum sint ab externis omnino affectionibus libera… The third passage, which is part of a result clause, is 990B-C …ὡς μηδὲ ταῖς αὑτῶν ἐθέλειν συγγίνεσθαι γυναιξίν, εἰ μὴ μύρων ὑμῖν ὀδωδυῖαι καὶ διαπασμάτων εἰς ταὐτὸ φoιτῷεν …ut non velint ad uxores suas accedere, nisi fuerint unguentis perfusae et odorum eos afflatus in idem coeuntium illexerint V in idem cocucium illexerint LC.Even after an emendation V was not able to give a satisfactory rendering of εἰς ταὐτὸ φoιτῷεν and the form cocucium of the other two codices does not make sense.What the Greek is talking about is that human males are not willing to go to bed with females unless they come covered with scents and perfumes to meet their lovers.The translator's original intention must have been concubium in keeping with the Greek phrase.In a similar context the word concubitum is used twice by Cassarino.Thus we would have: …nisi fuerint unguentis perfusae et odorum eos afflatus in idem concubium illexerint.
As mentioned in passing above, orthographic features may be capable of proving no less than scribal errors the existence of a common archetype.The anomalies, however, relating to orthography were removed to a large extent from V owing to Panormita's work of editorship.In spite of that, a few traces have still remained in the text to show that the codices go back to a common written copy.The words partly belong to the general vocabulary: 988D aeripides VLC < aeripedes; 989B diffiniam VLC < definiam; 990A olfatus VLC < olfactus; 990F aiuta VLC < adiuta; 992A adiscit VLC < addiscit; 992B adiscere VLC < addiscere.For obvious reasons, Greek proper names tend to keep their original form more frequently: 986A Heccatem VLC < Hecaten; Already in the passages discussed so far the reading of V occasionally diverged from LC.Now we are going to have a look at those numerous cases where Panormita made individual changes to the text inherited by the three codices (LC opposed to V).His sphere of operation ranges from correcting scribal errors to normalizing orthography, rectifying lapses of grammar, style, and meaning.Giving a full picture of Panormita's contribution to the text is not possible.The changes and corrections, wherever they seemed well founded, were adopted in our text and it would be useless to list them all.A limited number of instances taken from the several categories will suffice to illustrate his methods.
Among scribal errors corrected by V instances of omitting letters, replacing letters, and miswriting letters can be cited: 987A hic dicemus LC 43 habere dicemus V; 988A perdales LC pardales V; 988D deorum siles LC deorum similes Within the category of orthography a list of selected instances will do.The characteristic features shared by LC are simplification of double consonants, duplication of single consonants, and a variety of irregularities of aspiration and assimilation: 985E scenescentes belluarum LC Not all of Panormita's corrections were, however, successful.With some of the changes introduced by V we are left in uncertainty whether to accept or reject them.It seemed better not to follow them for the simple reason that LC have good readings to offer without a need for refinement.The few instances which belong here are related to certain aspects of grammar and style.These include the use of personal and demonstrative pronouns, conjunctive particles, and word order: 986A laus aut gloria LC laus et gloria V; 986C te omnes plane contempnimus LC omnes te plane contemnimus V; 987B sine magistro ac praeceptore LC sine magistro atque praeceptore V; 987F pro his quae genuerunt certamina LC pro iis quae genuerunt certamina V; 988D sic in his, qui bello clari sunt LC sic in iis, qui bello clari sunt V; 988E nec dentibus aut unguibus LC neque dentibus aut unguibus V; 991A in eiusmodi appetentiis LC in huiusmodi appetentiis V; 991A-B harum nos voluptatem semper LC harum nos semper voluptatem V; 992D calliditas ac celeritas LC calliditas et celeritas V.
We have reached a point in our enquiry now where new insights impose themselves upon us.They will come with the realization that Panormita's readings are not to be given preference over those of LC.The main evidence in support of this assertion is that Panormita made his changes to the text without knowing or consulting the original Greek.A fairly large number of passages cited here and further below will go to prove this. 45The editor of Cassarino's translation was not governed in his choices by what the original text had to say but allowed for the demands of style and meaning instead.
The most illustrative examples are given below.

45
Just as several of the isolated readings of V discussed so far may illustrate.
infinitos adhuc properes LC ut ego duco V.The choice of duco in V-if not due to scribal errormust have been dictated by Latin style rather than a knowledge of the Greek.There is no good reason for dismissing dico which is a literal rendering of φημί.
Ulysses is replying to Circe in 986A τί γὰρ δεῖ πoλλάκις ζυγoμαχεῖν ἡμᾶς περὶ τῶν αὐτῶν; Quid enim attinet de his rebus totiens adinvicem discrepare?LC disceptare V.The verb disceptare might have sounded less harsh in Panormita's ears than discrepare which is the original reading in LC.But the latter is more in accordance with the forceful meaning of ζυγoμαχεῖν in the Greek.
Ulysses is returning Gryllus' greetings in 986C Kαὶ σὺ, νὴ Δία, Γρύλλε.Et tu, inquam, Grylle.LC Et tu, inquam, salve, Grylle.V.The imperative salve corresponding to the Greek χαῖρε was supplied by V from the preceding sentence Xαῖρε, ᾿Oδυσσεῦ so as not to leave the reader with the impression that something is missing.
In some further cases, Panormita's doing the editorial work without the aid of the Greek has resulted in a misapprehension of the translator's original intention.This is especially true of those passages where Cassarino's translation is not as clear as one would like.
First, we have 989E …εἶχε δέ τι καὶ ἡ πόρπη χρυσὸς oὖσα παίγνιoν oἶμαι τoρείας διηκριβωμένoν …habebat enim aurum spinter quodam miro confectum artificio LC mirum V.This is a paraphrase of a famous passage in Homer with a description of a golden fibula Ulysses wore on his robe.V proposed reading mirum instead of miro.But miro should rather go with artificio as part of what was meant to be a rendering of τoρείας διηκριβωμένoν. 46 similar instance is 991D Eἶεν$ ἀλλ' ὄψoις χρῆσθε τoύτoις ἐφηδύνoντες τὴν τρoφήν$ Sic sane atque his tamquam ciborum condimentis utimini.LC utimur V.The verb form utimur in V is not compatible with the second person plural χρῆσθε of the Greek and therefore is wrong.It must be due to an improper understanding of the roles of interlocutors in this part of the dialogue.There is no change in the dramatis personae introduced with the concessive verb εἶεν.
One may finally cite 991D-E …oὐδὲ κoλλῶσα μελέτῃ καὶ συμπηγνύoυσα γλίσχρως τῶν θεωρημάτων ἕκαστoν πρὸς ἕκαστoν …nec scientiam scientiae conglutinans infirmatur, meditando confingit LC infirmat V.The double participle clause takes up an earlier feminine noun, animal φρόνησις, a faculty by which animals become experts in arts without having to join pieces of science together in a cumbersome way.The choice of an active transitive infirmat in V does not fit the context well and must be mistaken.We should rather take infirmatur as a curious attempt at rendering the Greek adverb γλίσχρως.
In a limited number of occurrences, the individual readings of V are more likely to be ascribed to scribal error.They can easily be checked by placing them side by side with the readings of LC: 986C sitis…, incideritis LC sitis…, inciditis V; 47 992D participarent LC parciparent V; 48 992E ratione et memoria LC ratione et memor V. 49 All pieces of evidence gathered so far must inevitably lead to the conclusion that codex V even though it has added valuable corrections to the text cannot lay claim to originality.Changes were made to the text of the common archetype by Panormita according to his own Latin taste 46 Panormita's confusion must have been caused by the absence of the Greek word παίγνιoν in the Latin translation.We should make an insertion like habebat enim aurum spinter <ludicrum> quodam miro confectum artificio.

47
The context requires that both verbs should be in the conjunctive.

48
Presumably as a result of a sort of haplography.

49
With a small letter a written above the last letter of memor.
and without recourse to the Greek.Next we shall have to turn our attention to LC and see if they can add to the picture about textual transmission by furnishing any detail.Our main concern will be whether these two codices differ in any way between them in their readings, and if they do, how V relates to these differences (L opposed to CV or C opposed to LV).
Insertions, omissions and word corruptions are the types or errors most suitable for establishing textual relations between codices and for ascertaining the dependence of one manuscript from the other.We should perhaps commence with word corruptions because they are relatively easy to trace and verify by means of the original text.The three following passages provide solid evidence for Panormita's dependence from L as well as for the priority of C over the other two codices.
The first passage is 991F …ἣν εἰ μὴ λόγoν oἴεσθε δεῖν μηδὲ φρόνησιν καλεῖν …quam si neque prudentiam nec rationem appellandam censetis C consentis LV.The second person singular verb consentis cannot be an accurate rendering of the second person plural verb oἴεσθε in the Greek.C alone gives the precise form censetis.A word corruption consentis < censetis must have taken place.
Similarly, we read in 992D …oὕτως oὐκ ἂν ἐδόκει ζῷoν ἕτερoν ἑτέρoυ τῷ φρoνεῖν ἀργότερoν εἶναι καὶ δυσμαθέστερoν …sic non videtur animal alio aliud imprudentius esse aut indocilius C iniocundius LV.The comparative iniocundius is completely out of place and must be the result of a banal scribal error.It is again C which alone offers a perfect reading.The corruption must have been iniocundius < indocilius.
Thirdly, there is 992E ᾿Aλλ' ὅρα, Γρύλλε, μὴ δεινὸν ᾖ καὶ βέβαιoν… Sed vide, Grylle, num verum ac certum illud sit… C vide, Grylle, nunc LV.The interrogative num in C, even if not a literal translation of the Greek μή after a verb of caution, is the only possible reading as nunc does not give good sense.The corruption clearly is nunc < num.
There is an extra passage with a severe text corruption in LV extending over several lines.An individual clause was removed from its original place by the scribe and later inserted in the wrong place.This resulted in a kind of jumble.The clauses still keep their original order (i-iv) in C: 992B …εὐμαθείας ἐπίδειξιν ὡς ἄλλo oὐδὲν oὐδαμῶς χρήσιμoν ἔχoυσιν.εἰ δὲ ἀπιστεῖς ὅτι τέχνας μανθάνoμεν, ἄκoυσoν ὅτι καὶ διδάσκoμεν.(i) …docilitatis scilicet argumentum, (ii) perinde nulla apud se potior sit cura.(iii) Si autem praeter fidem videtur nos artes addiscere, (iv) audi quod etiam docemus.C (i) …docilitatis scilicet argumentum.(iii) Si autem praeter fidem videtur vos artes addiscere, (iv 1 ) audi quod etiam (ii) perinde nulla apud se potior sit cura (iv 2 ) quod hactenus.LV. 50As is clear from the text of LV the clause (ii) perinde nulla apud se potior sit cura was inserted right in the middle of (iv) and what remained of the latter clause suffered corruption (hactenus < docemus).Another common error in LV not shared by C is the personal pronoun vos < nos.
Further proof of Panormita's strong reliance on L as against C can be adduced from scribal omissions and insertions.On two occasions, where L omitted a single word, empty spaces matching the number of letters of the missing part of the sentence were left in the manuscript as indicators that words of a particular length are missing.This procedure implies that the scribe not being able to make out the words in the copy before him wanted to supply them later.He failed to do so, however, except for a scribal note he wrote at a later 50 With a small cross indicating the problem in the right margin in V. date in the space allotted for the missing word in one of the two passages: 990C …oὐχὶ […] oἱ δ' ἄρρενες ὑπὸ oἴστρoυ καὶ μαργότητoς ὠνoύμενoι μισθῶν καὶ πόνoυ καὶ λατρείας τὸ τῆς γενέσεως ἔργoν …nec mares stimulo aut furore libidinis quasi oestro quodam allecti generationis opus pretio, labore aut servitute coemunt C pretio, labore aut |aut alia re| coemunt L 51 pretio, labore aut alia re coemunt V.The original reading servitute in C is a faithful rendering of the Greek λατρείας.It must have been illegible to the scribe of L who left the space empty in the first place, then filled the lacuna with a note saying there should be ʻsome other thing' in the series of ablatives.The text is about animal males not purchasing love with money, work, and service.All this was subsequently copied by V as if the words in the lacuna had been part of the text.
In another passage, the blank space L kept for the missing word was overlooked and skipped by V in the course of copying: 992D ὥσπερ εἰ σαυτῷ τὸν Πoλύφημoν ἢ τῷ πάππῳ σoυ τῷ Aὐτoλύκῳ τὸν Koρίνθιoν ἐκεῖνoν ῞Oμηρoν. 52Quemadmodum si te cum Polyphemo aut si avo tuo Ancolyco Homerum illum Corinthum conferes.C aut si |…| Anthiloco tuo L aut si Anthiloco tuo V.The word avo, of three characters, preserved in C is a rendering of the Greek πάππῳ.From the original word combination avo tuo the possessive adjective tuo was shifted to another position by the scribe of L after avo proved undecipherable.V copied the clause without taking heed of the lacuna.In addition, Anthiloco in LV reflects a more advanced stage of text corruption than Ancolyco in C. The respective stages must have been Ancolyco < Autolyco for C and Anthiloco < Autyloco < Autolyco for LV.
Two further instances of scribal omission in the text of L represent cases where words have simply dropped out.One passage is 986A ἐὰν δὲ μὴ πείσῃς, ἀλλὰ καὶ περιγένωνται διαλεγόμενoι… Sin autem nihil promoveris, sed te insuper rationibus revincent… C rationibus √ L 53 rationibus remittent V. Panormita's suggestion for reading remittent is only an attempt at saving the text of L. It misses the meaning.What Circe is talking about is the possibility that animals get the upper hand in the debate.The verb revincent in C is a perfect equivalent of the Greek περιγένωνται.
A slight change was made to the text in 991E καὶ μoυσικῆς ὅσoν ἑκάστῳ πρoσήκει κατὰ φύσιν …musicae autem quantum unicuique secundum naturam conveniat C musicae aurem √ unicuique LV.The change from autem to aurem in L must be due to the loss of the word quantum construed with the partitive genitive musicae.An 'ear for music' is ingenious but has no support in the Greek.
Of all evidence that can be cited in favour of the influence of L upon V the most compelling comes from casual interpolations.Any scribal addition in L which is not an integral part of the text, if repeated by V, can be used as a proof of the latter's dependence from the former.The first passage falling under this category is 988C ἵνα μὴ λάβῃ τραύματα ne vulnus accipiat C ne vulnus accipiat accipit LV.For some reasons, L has given two different forms of the same verb in succession, maybe as a result of an earlier copyist's superscribed correction from accipit to 51 I have marked the extent of the lacuna with vertical lines.The words aut alia re in the lacuna are written in a lighter colour which agrees with the colour of the letters used in the marginal notes.The first aut was probably underlined by the scribe after realizing that it had been inadvertently repeated in the lacuna.

52
The verb παραβάλῃς is to be supplied from the previous clause.

53
I have used a square root to mark the place of a word dropped out.
accipiat, later inserted in the text.It is the conjunctive form which is required by the Greek λάβῃ.Both verbal forms, however, were transcribed by V in imitation of its sample.
Another example is 991E τoὺς μὲν γὰρ Aἰγυπτίoυς πάντας ἰατρoὺς ἀκoύoμεν εἶναι… Omnes Aegyptios Medicos esse aiunt… C Omnes Aegyptios medicos Medicos esse aiunt… L Omnes Aegyptios medicos esse aiunt… V.This is an interesting case inasmuch as it shows again that the three codices had a common archetype not devoid of errors.The wrong interpretation 'Medes' can be excluded on the basis of the Greek word ἰατρoύς.The error was apparently recognized by the scribe of L, or his predecessor, who made an insertion with the correct form medicos. Then medicos was underlined as being the correct form by L and later copied with accuracy by V.
In a third passage, only a short insertion was made: 992A …ἀλλὰ κόρακας διαλέγεσθαι καὶ κύνας ἅλλεσθαι διὰ τρoχῶν περιφερoμένων …aut corvos loqui aut canes per circumagentes rotas saltare C per circumagentes sc.rotas saltare L per circumagentes et rotas saltare V.The abbreviation sc. may have been misread by V as an et.It is likely that L or an earlier scribe inserted scilicet, being unaware that circumagentes and rotas belonged together.A similar insertion of an explanatory nature in C in superscribed letters exerted no influence on the readings of LV in 992D oὐδενὶ γὰρ αὐτῶν ψυχῆς μέτεστιν nulli enim harum pars quaedam sc.sensus eius inest C nulli enim harum pars quaedam eius inest LV.
The evidence used in the instances of insertions, omissions, and word corruptions for establishing the connection between LV is scribal operation of a type which resulted in some form of deterioration of the text.There are a relatively great number of instances, however, when corrections leading to a better state of the text were initiated by the scribe of L. The most remarkable cases are furnished by those passages where traces of corrections left by the scribe's hand are visible in the manuscript: 987A videar C videat L 54 videat V; 989C molestia C modestia L modestia V; 55 990D ade brevis ac tenuis C adeo L 56 adeo V; 990D fugientes C fugientem L 57 fugientem V; 991C veracitatem C voracitatem L 58 voracitatem V.It is worth noting that in each of the passages cited the old readings of C and L must go back to the common archetype as on the testimony of the codices we should not suppose that any direct connection existed between C and L. To make the list complete, one should also add the specific instance of 989C deffensoribus C deffectionibus L 59 defectionibus V.The problems of this passage have already been dealt with in full.
Other corrections made by L to the text of C need not detain us here.They are for the most part concerned with scribal errors of the usual type, including miswriting or dropping letters and words, and peculiarities of orthography. 60Some doubtful readings might be settled by relying on the authority of the Greek, like the hesitation between singular and plural cases: 986B πῶς γὰρ  The relative cum, for instance, has a tendency to be written as quom in C, but not in all parts of the text.ἂν ἢ δoῖεν oὗτoι λόγoν ἢ λάβoιεν… Quonam enim modo hi sermones intellexerint aut ceperint… C hi sermonem intellexerint L sermonem hi intellexerint V; 61 987B …πρὸς γένεσιν ἀρετῆς …ad generationem virtutum C virtutis LV.Otherwise, it is not always easy to decide whether or not the errors themselves originated with the common archetype. 62t would, however, be unwise to subscribe without qualification to every single change urged by L. With some of the corrections, the scribe in his general attempt at standardization risked wiping out original traits of the translation.One instance is 986C ὡς μάτην ἄρα δεινὸς ἐλέγoυ quod falso prudens dicebare C dicebaris LV.The alternate ending -re in the second singular passive is not unusual in Cassarino who beside the -ris forms used a present conjunctive appellere as well as an imperfect indicative persequebare in the same dialogue.Nor can a replacement of an adverb be defended in 991F ὥσπερ ἀμέλει καὶ δι' ἔργων… Ex operibus quippe… C Ex operibus enim… LV.The adverb quippe in the same role is used more than once by Cassarino in the dialogue.
Two small groups show slight changes which do not seem necessary.We should rather follow the text of C in the use of certain conjunctive particles: 989B vanas ac superfluas C vanas & superfluas L vanas et superfluas V; 63 991E quis autem docuit C quis aut docuit LV; 64 991F neque prudentiam nec rationem C neque prudentiam neque rationem LV; 65 992A lasciviam aut ludum C lasciviam atque ludum LV. 66 Another group makes change in the word order.Here, again, the readings of C should be preferred: 67 991F convenientius ac dignius C dignius ac convenientius LV; 992C in hominum manibus C in manibus hominum LV; 992D animal alio aliud imprudentius C animal aliud alio imprudentius LV; 992E verum ac certum illud sit C verum ac certum sit illud LV.In the same sequence, even a blatant error was made, later recognized and restored by Panormita: 992C sine ratione existimarem CV sine existimarem ratione L. 68 What remains to be answered is whether there is any possible relation of dependence between C and V.The answer must be given in the negative.Wherever CV share in a common reading it is because L goes astray in the text.This means orthographic irregularities of the kind which may well go back again to the common archetype.By far the most frequent errors are duplication of single consonants and simplification of double consonants.They might as well be the fault of the scribe of L. So the agreement between C and V in these passages is illusory.
To sum up the results, the three codices CLV had a common archetype [α], already at some distance from the translator's original copy.CLV inherited a considerable number of scribal errors from [α].C stands closest to this common archetype and represents a separate branch.At the same level of another branch stands [β], the ancestor of L. In addition to the common stock 61 With change in the word order in V.

62
It is equally possible that neither C nor L corrected an error, either C or L corrected an error, both C and L corrected an error.In addition, either or both scribes can bring their own individual errors in the text.

63
The abbreviation sign & not being unequivocal was resolved as an et by V.

64
The Greek has a particle δέ.

65
Asymmetric constructions like neque… nec are not uncommon in Cassarino.

66
The Greek has a particle ἤ.

67
Curiously enough, all cited examples come from the end of the dialogue.

68
The words sine and ratione closely belong together.of errors derived from [α], L inherited a further number of text corruptions from [β], not present in C. V strongly depended on L and tried to do its best to correct the errors with no aid from the Greek.The different stages of textual transmission and the genealogical relationship between the codices are represented in the form of a stemma below: It is beyond doubt that the codex of the Vatican Library belonged to Panormita.At the end of the translations of the Plutarchean works a short note reads Antonii Panhormitae est and the ownership is confirmed by an ex libris on one of the last flyleaves of the codex. 69The manuscript together with a number of other codices once owned by Panormita found its way into Fulvio Orsini's library 70 and was in turn acquired by the Vatican Library as part of a larger collection of Latin manuscripts in 1658.There may be some uncertainty as to whether the codex is an autograph.Scholars refraining from making positive statements do not seem to be much inclined to admit the possibility. 71It seems, however, very likely that Panormita was the editor and copyist in one person.The handwriting in the main text and the marginal notes, occurring here and there in the translations, 72 display similarities to an extent which would make it hard to believe that they do not come from the same hand. 73That the marginal notes are attributable to Panormita is an established fact.Even from the one single genuine note added to the text of the Bruta animalia Panormita's identity becomes evident. 74When Gryllus is talking about the land of the Cyclopes, well known to Ulysses, endowed with the capacity to produce crops spontaneously without being ploughed or sown, Panormita's short note specifies the place as 'Sicily'.As a means for authentication of Panormita's handwriting one can use three of his codices, all acknowledged as autographs, which contain elegies in Latin and a collection of letters sent to and received from distinguished humanists. 75n exact knowledge of the provenance of the manuscript of the Biblioteca della Società Siciliana per la Storia Patria would certainly help us define the place we should assign to this particular codex which, at any rate, must have been the direct source of Panormita in reconstructing the text of Cassarino's translation of the Bruta animalia.At the time of Cassarino's death and in the following years Panormita was living at the Neapolitan court of Alfonso V of Aragon, the place where Cassarino himself wanted to move in hope of a better living before his life tragically ended.And yet it may be of some significance that we find a codex entirely devoted to Cassarino's translations in a library of Panormita's native town, Palermo. 76he details as to how the manuscript came into the possession of Giuseppe Lodi who donated it to the library are unknown.The handwriting is quite different from Panormita's and the scribe who made use of Latin abbreviations more often than usual has left a large number of notes in the margins.These marginal notes which consist of a series of keywords and brief phrases reformulating the contents attest to a thorough reading of the text.
The codex of the Biblioteca Casanatense of Rome gives the best and most authentic reading extant of Cassarino's translation of the Bruta animalia.The manuscript, composed of a variety of miscellaneous works with no manifest link between them at first sight, has only preserved two of Cassarino's translations, Bruta animalia ratione uti and Regum et imperatorum apophthegmata.The reason for the inclusion of these two translations in the same manuscript may be sought for in the person of the dedicatee.Both of the translations as we have them in V 77 are dedicated to Cassarino's disciple Giacomo Curlo. 78It would be tempting to see the presence of these two works in the same manuscript as somehow being connected with Giacomo Curlo's person.A twist comes with C, however, where in place of Giacomo Curlo's name Balbi is given as addressee of the dedicatory letter of Regum et imperatorum apophthegmata.The first line of the text goes: Nuper, amoenissime Iacobe, aliquantulum otii nactus cum essem… V; Nuper, amoenissime Balbe, aliquantulum otii… C. How this replacement has taken place is a riddle. 79ll that we know about Balbi, an expert of both Latin and Greek, can be gathered from Cassarino's advice for Giacomo Curlo in the dedicatory letter of Bruta animalia that he should be careful not to share the translation with anyone before consulting Balbi. 80There is a similar mode of expression which may give us a clue in the text of the dedicatory letter preceding the Regum et imperatorum apophthegmata.Here Cassarino is referring to a former occasion when he had sent his translation of the Apophthegmata Laconica to his addressee for consultation before passing it over to anyone for reproduction. 81These words are better suited to be addressed to Balbi, a qualified expert, than to Giacomo Curlo, an aspiring student of Cassarino.
Not incidentally, with the Codex Casanatensis we are in a Genoese milieu of the middle of the 15th century.One of the copyists of Leonardo Bruni's writings in the codex left his signature as Schiaffino da Camogli the Genoese with a date 1446 in Genoa.Furthermore, the manuscript ends with three interesting pieces, each written on Genoa or Liguria, conjuring up figures known 76 An alternative explanation for the emergence of the codex in Palermo would be Cassarino's Sicilian origin.from Cassarino's personal acquaintances such as Giacomo Bracelli and Andrea Bartolomeo Imperiali. 82It remains to be seen whether the oration at the end of the codex, of an unknown author, is a posthumous edition of Cassarino's public speech otherwise known to have existed. 83

THE GREEK SOURCE TEXT OF CASSARINO'S TRANSLATION
There are over thirty Greek codices known to date in which the dialogue Bruta animalia ratione uti is extant.These are distributed among the various families and branches which, according to general scholarly consensus, constitute the genealogy of the textual tradition of Plutarch's Moralia. 84Whereas the codices of family Π, after they had been compiled by Maximos Planudes and his circle in the end of the 13th century, exerted much influence upon the formation of the text of all other families in general, the measure of this influence was twice as much in family Θ as in family Ψ or in its related branch Ξ. 85 What facilitates the task of specifying the Greek source text of Cassarino's translation is the fact that he used a codex for his translation which belonged to family Ψ, namely codex g. 86 The close agreements between Cassarino's translation and several of the readings of codex g not attested anywhere else in manuscript tradition will give incontestable proof of this statement.Some of the isolated readings in g have single words omitted.They are without exception followed by Cassarino: 986D …μή σε πoιήσῃ [λαθoῦσα Ω] 87 σῦν ἢ λύκoν …ne te aut suem aut lupum faciat; 88 990A ἡ μὲν γὰρ γλῶττα τoῦ γλυκέoς [καὶ δριμέoς Ω] καὶ αὐστηρoῦ γνώμων ἐστί… Lingua quippe dulce ab amaro internoscit…; 991A …ὡς δ' ἐγᾦμαι καὶ Σφίγγες ἀναβλαστάνoυσι [καὶ Kένταυρoι Ω] …et, ut ego puto, Sphinges nascuntur.
In a few instances, g has a reading which diverges widely from the text of the other codices.Cassarino's translation in each of the occurrences follows its model closely and provides 82 An Italian poem with the title De Genua urbe (121 v -123 r ) is followed by Giacomo Bracelli's letter Descriptio orae Ligusticae written in 1448 (123 v -127 v ) as well as an oration Laudes urbis Genuae of an unknown author (127 v -135 v ).The main codices belonging to the individual families or branches are αAEG for Π, ZυB for Θ, Pq and gy for Ψ, and Qihk for Ξ.

87
Missing words in g are put in square brackets in the text with Ω indicating the reading of all other codices except g.

88
In another passage of the dialogue the Greek participle λαθών is translated by Cassarino as latens.
indisputable evidence that it was based on a variant including all these isolated readings: 986E ἀλλὰ μὴν ἐγὼ πρόθυμoς (πρὸς ὑμᾶς Ω) 89 ἀκρoᾶσθαι Sum equidem ad audiendum paratus; 90 990F καὶ γὰρ αἰγῶν ἐπειράθησαν ἄνδρες καὶ oἰῶν (ὑῶν Ω) καὶ ἵππων μιγνύμενoι… Nam hominum plerique cum pecudibus, capris aut equabus coiere…; 992E ἀλλ' ὅρα, Γρύλλε, μὴ δεινὸν ᾖ καὶ βέβαιoν (βίαιoν Ω)… Sed vide, Grylle, num verum ac certum illud sit… All of the text omissions and individual readings of codex g, when juxtaposed with Cassarino's translation, show clearly that the source text Cassarino used for his translation was g or a similar copy.It can also be asserted that he used no other source for his translation except g since he left all passages as they are in g and introduced no change in any of them.Some doubtful cases may be easily explained by the translator's ability to find out a better reading.One example will suffice.In all manuscripts including g the passage 987D reads oὐδὲ δoυλεύει λέων λέoντι καὶ ἵππoς ἵππῳ δι' ἀνδρείαν, meaning that neither a lion nor a horse will suffer slavery from another of the same breed owing to the virtue of courage.A variant reading δι' ἀνανδρίαν was, however, introduced by a second hand in J, later adopted by editors, with a change in the meaning that animals will not suffer slavery by cowardice.Cassarino appears to have followed the same line of thought when translating nec leo leoni nec equus equo ob ignaviam servit.
Little is known about the fortune of codex g.Ten years after Cassarino's death it was given to Giovanni Aurispa by the Florentine Francesco Biliotti in return for a number of other books borrowed from him.The fact of the exchange was recorded on a flyleaf of the codex by the owner in Rome, 1457.A previous owner's name Andrea di Biagio is given by an ex libris on the same page. 91Cassarino may have had access to the manuscript for a time in the years between 1440 and 1445 to make his own copy of the dialogue Bruta animalia ratione uti.The codex does not, however, contain other works of Plutarch translated by Cassarino except for Regum et imperatorum apophthegmata and Apophthegmata Laconica.Whether it was the source used for the translation of the latter two works is a question left for future study.

89
The readings of the other codices are marked with Ω and put between brackets.90 Codex P is to be added which records πρόθυμoς written by a second hand.91 Vat.Pal.gr.170, 327 r .An identical exchange note together with the same ex libris is also to be found in Vat.Pal.gr.176 and 192.Cf.STEPHENSON, H.: Codices Manuscripti Palatini Graeci Bibliothecae Vaticanae.Roma 1885, 89-90; MAN-FREDINI, M.: Codici plutarchei di umanisti italiani.Annali della Scuola Normale Superiore di Pisa 17 (1987) 1034, n. 190.   92   The present edition of the text of Antonio Cassarino's Latin translation of Plutarch's Bruta animalia ratione uti is largely based on the readings offered by the codex Biblioteca Casanatense 665 (C II 8), Rome, designated here and elsewhere as C. Some of the readings of the other two codices, MS Lodi XII E 13 of the Biblioteca della Società Siciliana per la Storia Patria, Palermo, and Vat.lat.3349 of the Biblioteca Apostolica Vaticana, signed L and V respectively, were only introduced in the text where it seemed necessary in order to avoid scribal errors represented by C. For convenience's sake, folio numbers of all three codices are given throughout in brackets, in addition to the reference numbers put in square brackets of the original Greek text of Plutarch's Moralia.The text follows orthographic standards of Classical Latin rather than Humanist usage.A careful interpunctuation was added with a view to better reading.nec otii mihi satis (nam occupationes meas nosti) ad id quod volebam suppeteret, disputationem hanc brevissima lucubratiuncula e Plutarcho tibi transtuli, non quod animo meo id satis crederem, sed ut intelligeres quam nihil mihi studio tuo sit gratius et quam genere in hoc vel parva etiam soleam consectari.Tu cave prius cuiquam hanc communices quam Balbum nostrum utriusque linguae doctissimum consulueris.Eius enim auxilio poteris non modo si qua minus commode dicta erunt emendare, sed si qua etiam obscuriora videbuntur intelligere.

40
Was it a gloss to supplement the meaning of salvos?41 With evident signs of correction from deffensionibus to deffectionibus.42Withtraceable signs of deletion of an earlier deffensionibus by rubbing.

54
With visible signs of an earlier r being corrected to t.55With visible signs of an earlier l being corrected to d in L as well as V.The word molestia does appear at the end of the previous sentence as a potential cause of the scribal error.56 With a large letter o written above e.57 With the last letter s erased and a nasal suspension for m presumably added.58 With visible signs of an earlier e being corrected to o. 59 With an earlier deffensionibus in C as well as L corrected to deffensoribus in C and to deffectionibus in L. 60 , P.: La bibliothèque de Fulvio Orsini.Contributions à l'histoire des collections d'Italie et à l'étude de la renaissance.Paris 1887, 218-223.The codices are of three or four categories: Latin classics, humanist translations from Greek, manuscripts of high antiquity, and codices related to Panormita's life and works.71 DE NOLHAC (n.70) 221; RESTA (n.23) 227-228: "La scrittura è molto vicina a quella del Panormita, però non è sua"; INDELLI: Traduzioni (n. 3) 230-232.72Themarginal notes are more numerous in the last two works Regum et imperatorum apophthegmata and Apophthegmata Laconica than elsewhere.73Thishand is also to be recognized in Vat.lat.3346, 1 r -26 v .74Foran enumeration of the notes present in the dialogue, seeINDELLI: Traduzioni (n. 3) 232, n. 45.   75   Vat.lat.3273, 3371-3372.The latter two codices which contain letters prepared for publication show a more cursive character of handwriting as compared to the former codex of Latin elegies.