Justin Stover has recently edited a collection of Platonic placita, organized by individual dialogue, which he identifies as the lost third book of Apuleius’ De Platone.Footnote 2 The work is preserved only in a thirteenth-century manuscript, Vatican BAV Reg. lat. 1572 (= R). The manuscript is filled with trivial errors, including a large number of one-word or two-word lacunae. Stover has worked ably to clean up the text and many of his emendations are uncontroversial. But any editio princeps is likely to be susceptible of improvement. I offer here some notes on a few points where I think either the text or Stover's translation and commentary can be bettered. I am not primarily concerned here with the ascription to Apuleius (if the work is not by Apuleius, it must in any case be by a Middle Platonist roughly contemporary with him), although one or two points may have some relevance to this question.Footnote 3
It may be helpful to note at the outset that entries in the work commonly take one of two forms: a sentence with Plato as understood subject (‘he says, argues, shows …’, etc.), or an accusative and infinitive construction motivated by an understood (and sometimes explicitly expressed) placet illi. The reader should be warned also that, while Stover has done yeoman work in tracing the individual placita to their source-passages, the author's use of Plato is so vague and wilful that the Greek model is not always of much help in establishing the text of the Latin.
The translations appended to the passages are Stover's.
1.15 (p. 96)
placet illi maiores natu cum eo tamen ut optimi sint honoribus fungi.
He is of the opinion that those who are greater by birth should receive honours, provided that they are the best.
cum eo tamen ut … (‘with the proviso that …’) seems to be found elsewhere only in Celsus (3.5.10, 4.13.5, 7.2.5; cf. 1.3.21 cum eo tamen ne …, 2.17.4) and Africanus, Dig. 46.3.39 (cited by Stover per litteras). Celsus also (and in fact more frequently) has cum eo ut without tamen: see the examples cited by B. Löfstedt, ‘Notizen zu Sprache und Text von Celsus, De medicina’, MH 47 (1990), 60–2, at 61. For cum eo ut Löfstedt cites also Liv. 8.14.2, 8.14.8, 36.5.3. This makes one wonder whether R's et should be retained but bracketed, as a false anticipation of ut.
3.20 (p. 100)
improbat deinde eos qui negantis homines in seruitute habeant aut qui omnino eiusdem ciuitatis nationem belli iure diruant aut qui hostium spolia deorum aedibus adfigant.
He then rebukes those who hold people in slavery against their will, or else who destroy utterly the people of that same city by right of war, or who hang the spoils of enemies on the shrines of the gods.
Most, if not all, slaves are enslaved unwillingly, so it is not clear what negantis would add. Plato's prohibition, moreover, is not on enslavement per se but on enslaving other Greeks (Rep. 5.469b8–c7). Also striking is the alternative accusative ending in –is, found elsewhere only at 3.16 (p. 100) and 18.13 (p. 114) deos caelestis, where archaism seems more in place. I suggest that negantis conceals a term equivalent to Graecos, which in turn would help explain eiusdem ciuitatis. Something like cognatos or <eiusdem> nationis would fit for sense, but I cannot find anything that seems immediately right and would therefore obelize for the time being. (Also odd is eiusdem ciuitatis nationem; would one not rather expect eiusdem nationis ciuitatem?)
3.24 (p. 100)
praeterea philosophari <nihil> aliud quam studere sapientiae; cuius desiderio philosophantes teneri.
Furthermore, to practise philosophy is <nothing> other than to be zealous for wisdom; by the desire for it, people are deemed philosophers.
Rather: ‘those who philosophize are gripped by desire for it.’ Cf. (for example) Cic. Sen. 33 magno desiderio tenebitur uirium; Plin. Ep. 7.5.1 incredibile est, quanto desiderio tui tenear; Ambrose, Hex. 6.4.25 quamuis cibi desiderio teneatur et potus.
3.29 (p. 100)
natura uero esse aliquid iustum, honestum, bonum, legitimum existimat, substantiam incorporalem habens; etiam solum bonum esse quod honestum est.
He holds that by nature something just, decorous, good, and lawful exists, and it has incorporeal being. Also, the only good is what is decorous.
The note on p. 144 reveals that R reads not just habenciam but habenciam et. The author does not typically begin a new placitum with etiam; the only example I have found is 26.11 (p. 122). He often begins one with et, for example 17.8 (p. 114); 21.11 (p. 118); 23.1 (p. 120); 25.3 (p. 122); 26.2 (p. 122). Note especially the parallel passage at 23.3 (p. 120) et solum honestum quod bonum. I think it is easier to retain et and explain habenciam as a perseveration error after substantiam.
4.8 (p. 100)
denique in homine uirtutem consummari tunc cum ingenio ante institutionem quoque rectam et honestam uoluntas adiutauerit dei.
Then, virtue is brought to fulfilment in a man, at the point when the will of god has assisted his talent even before upright and decorous instruction.
Stover notes that ‘adiuuare with a dative is rare and late’, citing TLL 1.718.72–8, which gives examples from the Vetus Latina and from fourth-century authors. Hence his adiutauerit, ‘a good Apuleian word, and one capable of taking the dative (cf. Petron. Sat. 62.11)’. Adiutare + dat. is also found in the letters of Claudius Terentianus: cf. J.N. Adams, The Vulgar Latin of the Letters of Claudius Terentianus (Manchester, 1977), 42. As Adams notes, the construction is obviously modelled on other verbs of helping (for example succurrere, auxilium ferre). The conjunction of Terentianus and Petronius (in a freedman's speech) suggests that the usage is a vulgarism and their weight as parallels for our text is correspondingly diminished. Better to print the (virtual) paradosis adiuuerit, and accept that this may be the earliest example of the construction. Our author may have been influenced here by Rep. 6.492a ἐὰν μή τις αὐτῇ βοηθήσας θεῶν τύχῃ. For what it is worth, adiuuare is also ‘a good Apuleian word’, and in fact more frequent in Apuleius than adiutare.
7.4 (p. 102)
sic effici ut ad omnes bonos beatitudo pertineat, ad malos miseria.
So it comes about that happiness pertains to all the good and misery to the wicked.
The other occurrences of this phrase (8.8; 16.6; 22.10) are translated similarly. Better would be: ‘Things are so arranged that …’. Our author seems to have a fondness for this rare idiom, and if the author is Apuleius it is curious that it does not appear elsewhere in his works. Stover per litteras points to Apol. 68.1 p. 82 Vallette ordinem … persequor et efficio ut …; Flor. 9.34 p. 139 Vallette qua effec<is>ti ut (Van der Vliet; effectius F) te amare debeant experti. But these are not quite parallel.
8.19 (p. 106)
denique concedit fieri aliqua fato et necessitate, rursus esse aliquid in nobis quod sustineatur bona ratione uiuendi uel mala.
Finally, he grants that some things happen by fate and necessity. In turn, there is something in us which is sustained by either a good or an evil mode of living.
The meaning of Stover's sustineatur, ‘is sustained’, could probably be extracted from the transmitted contineatur (cf. OLD contineo 3, ‘To keep or hold together …, prevent from disintegrating, sustain’; also 4 ‘To maintain or keep … in a specified state or condition’). I wonder, however, if contineatur does not mean ‘is dependent on, is conditioned by’ (OLD contineo 12b; TLL 4.707.15–53). This would make a better counterpart to fieri … fato. Some things are fated, some depend on us and on the kind of life we choose to lead.
12.1 (p. 108)
in Critone decernit somnia quoque fieri nec officii esse cum ualitudinario corpore diu uiuere.
In the Crito, he also lays down the doctrine that dreams happen and that it is not one's duty to live for a long time in a sickly body.
That people dream is uncontroversial, so it is hardly a ‘doctrine’ that needs ‘laying down’. fieri, I think, means not ‘happen’ but ‘come to pass’, ‘be realized’. Socrates dreams that he will be executed on the third day (Cri. 44a1–b4), and so indeed it proves.
13.2 (p. 108)
[In the Phaedo] aeque de somniis loquitur tamquam de prouidentia deorum uetante excedere †autem† uita; et uoluntaria morte obire nulla ratione dictante non iudicat.
… he speaks both about dreams and about the providence of the gods, which forbids one to depart from life † … † and he does not think it right to undergo voluntary death, if no reason requires it.
Clearly autem is corrupt. Stover suggests that it conceals a Greek term meaning ‘suicide’, but the author does not tend to scatter unglossed Greek around like this.Footnote 4 More likely some word meaning ‘prematurely’ or ‘unbidden’. Perhaps temere (for which, compare Val. Max. 2.6.7 egredi uita temere, in a discussion of suicide). Still easier might be ante <die>m, ‘before our (appointed) day’, a well-attested phrase used famously of Dido's suicide at Verg. Aen. 4.697.
13.7 (p. 108)
sensus non putat esse subtiles nec in totum uerum ueros sed de rebus iudicare solam posse rationem.
He does not think the senses are subtle, nor accurate for the whole truth, but only reason is able to make judgements about things.
Stover notes that ‘I have followed the manuscript here, even though its meaning is a bit murky’, and indeed both the English and the Latin seem very awkward. (Phd. 65b1–8 is no help.) For in totum uerum Stover cites Sen. Ep. 92.3 si ueritas tota perspecta est, which is really no parallel. I should prefer to delete uerum as the scribe's first, botched, attempt at ueros, led astray by the preceding totum. Adverbial in totum = ‘entirely’, as at, for example, Celsus 1.3.39 Venus … aestate in totum, si fieri potest, abstinendum est; Sen. De ira 1.17.7 in totum inaequalis est. Apuleius has the phrase at Plat. 2.9 p. 121.9–11 Moreschini non illas quidem disciplinas uocandas esse censet, sed non in totum existimat disciplinis alienas.
13.24 (p. 110)
terram deinde existimat formam habere rotundam …; habitare etiam homines in <de>pressis eius et raris regionibus.
Then he holds that the earth has a round form … People live in its hollows and in widely dispersed areas.
According to Phd. 109a9–b7 (the famous ‘frogs around a pond’ passage), Socrates believes that the world is heavily populated. So it is disconcerting to be told that men live ‘in widely dispersed areas’. Stover's introduction (82) notes a potential parallel at Cic. Somn. 20: uides habitari in terra raris et angustis in locis. But in that passage Scipio is making a quite different (indeed, opposite) claim, that the inhabited regions of the earth are a small proportion of the total. Plato's Socrates, by contrast, tells us that there are πανταχῇ περὶ τὴν γῆν πολλὰ κοῖλα καὶ παντοδαπά, where men can live (109b4–5). The second adjective would be well rendered not by raris but by uariis; the nonsensical rarus with its two minims preserves traces of the true reading.
15.5 (p. 110)
diuisionem deinde bonorum facit in duas partes: prima itaque et diuina bona ait esse uirtutes, secunda autem et humana ea quae uulgo bona putantur, sed <…> si secundum uirtutem usus sit illorum.
He then makes a division of good things into two parts: the primary and divine goods he says are the virtues, while the secondary and human are those things which are commonly considered good. But <…> if their use is in accordance with virtue.
Stover does not venture to fill the lacuna, but <tantum> si or (as A.J. Woodman suggests) <non ni>si would restore sense. This is the positive counterpart of 10.3 ostendit … quomodo ea quae uulgo bona existimantur, si non cum uirtute habeantur, nullius pretii sint.
15.10 (p. 112)
nec permittit uoluptatis causa coire, sed tantum liberorum procreandorum.
He does not permit sex for the sake of pleasure, but only for the procreation of children.
The transmitted creandorum can stand. Cf. e.g. Val. Max. 7.7.4 non creandorum liberorum causa coniugium intercesserat; Quint. Decl. min. 247.6 coisse … liberorum creandorum gratia; Lact. Epit. 57.3 libido extra legitimum torum non euagetur, sed creandis liberis seruiat.
16.2 (p. 112)
partem praeterea eruditionis esse honeste cantare et saltare; neque musicae enim finem positum in uoluptate quia solum sit bonum quod honestum et solum malum quod turpe.
Furthermore, part of education is to sing and dance in a decorous manner, and the end of music is not truly located in pleasure, since the only good thing is what is decorous and the only evil is what is shameful.
The placement of enim is odd, and it is difficult to extract Stover's ‘truly’ from it. As the referee points out, his rendering would better fit <u>erum (adj.)—certainly an attractive suggestion. Alternatively, we might simply delete enim as an anticipation of finem; for neque introducing a new doctrine, cf. 22.8 (p. 120) neque inter stultos amicitiam esse. Or (as A.J. Woodman suggests) we might transpose musicae and enim, and translate the latter normally. Singing and dancing should be taught, but only decorous singing and dancing, for the end of music is not pleasure (but virtue).
16.9 (p. 112)
nec uero expedire stulto diu uiuere, nec multo magis expedire alia quae uulgo bona existimantur quam mala.
And it does not benefit a fool to live for a long time, and those things which are commonly held good do not benefit him any more than those thought evil.
Stover's translation seems to reflect illi ea rather than his own alia, and I am not sure that the latter is needed. The rendering also ignores multo, which is indeed problematic. At Leg. 2.661a4–b7 the Athenian says that things commonly thought good (such as health and a keen eye) are of benefit to the good but harmful to the bad—not that they are ‘only marginally better’ than bad things. We might be tempted to delete multo, but the phrase is supported by 7.14 et multo magis uitam prudentis hominis uoluptate frui quam imprudentis; 19.18 et multo magis cum uoluptate uiuere sapientem quam stultum. It may be that the epitomator is simply being sloppy here.
16.14 (p. 112)
intra duodeuiginti autem annos agenti aetatem <uinum> uniuersum denegat.
To those under eighteen years of age he denies wine entirely.
Clearly we need a word for ‘wine’, and uinum could easily have fallen out before uniu-. But uniuersum is also odd (as ‘the whole of wine’ would be in English); one would expect omnino. I initially suspected an abbreviation error, the scribe having mistaken uim for ūnm (i.e. uinum for uniuersum). But Stover per litteras reasonably directs me to Leg. 2.666a3–5: τοὺς παῖδας μέχρι ἐτῶν ὀκτωκαίδεκα τὸ παράπαν οἴνου μὴ γεύεσθαι. This speaks for the authenticity of uniuersum, but the objection to the idiom remains. I therefore propose <uinum in> uniuersum, which would make for an even easier haplography than uinum alone.
18.3 (p. 114)
nullius alioquin pretii esse uitam hominis si non id agat <ut> meliorem se ipse adsidue †exhibere†.
Then, the life of a man has no value if he does not take care to † … †.
Stover's restoration of ut seems inevitable. He leaves ‘the bizarre and out-of-place exhibere’ in place. One might simply emend to exhibeat (with the referee), but the corruption of the ending is not immediately explicable. The text is full of brief lacunas, so perhaps <studeat> exhibere? A.J. Woodman suggests ex<erceatur ex>hibere.
18.13 (p. 114)
religionis ordinem ita disponit ut primo deos caelestis colamus, deinde infernos.
He arranges the order of religion such that we worship the heavenly gods first, then the gods of the underworld.
The author normally speaks of Plato in the authorial present, but is disposuit (‘has disposed’) really impossible? A similar case is 23.10 (p. 120):
apud inferos quoque imperium animas puniendi deorum prouidentiae delegat.
He also assigns the authority to punish souls in the underworld to the providence of the gods.
delegat is some distance from dedit, and the latter does not seem inconceivable. For imperium dare, cf. TLL 5.1.1676.33–41 and add, for example, Plin. Pan. 38.7. (By contrast, I have found no parallels for imperium delegare.) I also wonder if animas puniri is really insupportable (‘authority for souls to be punished’); if it is, would punire be an easier change?
18.17 (p. 114)
constituit et quot annorum quisque ducere uxorem debeat et idoneam aetatem esse existimat quae a triginta annis usque ad quinque amplius ducendi spatium habeat.
He establishes also at what age each man ought to take a wife and says that the proper age is that which affords a space for marrying from the age of thirty to five more.
This is oddly verbose, and a triginta annis usque ad quinque amplius is a very unnatural way of saying ‘from thirty to thirty-five’. (Contrast 3.10 usque ad annum quinquagesimum quintum; 21.1 a quinquennio usque ad uicesimum annum.) Perhaps a few words have been skipped in copying, something like:
… et idoneam aetatem esse existimat quae a triginta annis usque ad quinque <et triginta post labores> alios ducendi spatium habeat.
21.5 (p. 118)
probat autem eos maxime qui neque uoluptates magnopere curent et doloribus patientiam praestent dumtaxat ratione dictante.
But he approves most of all of those who do not care greatly for pleasure and who have patience in the face of pain, as his reasoning requires.
The phrase in bold means rather ‘provided that reason dictates it’. Displays of endurance that stem from mere bravado are not approved. Cf. 13.3 (p. 108) uoluntaria morte obire nulla ratione dictante non iudicat (where Stover correctly renders ‘if no reason requires it’).
21.7 (p. 118)
deinde morem quid sit et ait legem non inscriptam.
Then [he explores] what custom is and says that it is unwritten law.
The lack of a verb in the Latin is troubling; it cannot be supplied from the preceding sentence (for which, see the previous note). Contrast 7.18 (p. 104) quaerit deinde quid sit suum negotium agere et ostendit …; also 15.15 (p. 112) definit post hoc quid sit eruditio et ait institutionem esse eam …. Should we read, for example, quid sit <quaerit>?
For legem non inscriptam Stover compares Isid. Orig. 2.10.1 mos uero est … lex non scripta. The last three words appear to be the normal Latin equivalent of ἄγραφος νόμος: Cic. Mil. 10 non scripta sed nata lex; Tert. Adv. Iud. 2.7 legem … non scriptam; Ambrose, Hex. 5.21.68 leges naturae non scriptae litteris. I have found no parallels for lex non inscripta—a phrase that would court confusion, since (as A.J. Woodman reminds me) inscriptus is used elsewhere to mean ‘unwritten’: cf. TLL 7.1.1850.65–80 and note CGL 3.452.32 inscriptus: ἄγραφος (sim. 2.217.9; 3.483.36). It is tempting to delete non, which might have arisen as a gloss on in-, or crept in (as ñ) between -m in-. On the other hand, inscribere is occasionally found with the force of scribere (TLL 7.1.1848.57–70), and defenders of the transmitted text might appeal to Epistula Barnabae 7.3 inscripta lege praecepti (= γεγραμμένης ἐντολῆς).
25.7 (p. 122)
improbat et causidicat artificium tanquam malum et eos qui callide agant causas morte iudicat puniendos.
He blames and indicts artifice as wicked and he decrees that those who argue cases with trickery should be punished with death.
Despite Stover's defence of it, causidico(r) has only the most tenuous claim to existence, and the metaphor would be a drastic departure from our author's neutral style. It would be especially perverse to depict Plato himself as a lawyer in a context where he is criticizing lawyers. Stover reluctantly suggests causidicarum as a possible alternative; this seems much preferable.
27.12 (p. 124)
in hoc libro adicit quintum corpus quod aethera uocat, ita <ut> cum hoc prima corpora sint ignis, aer, aqua, terra. et in his omnibus uiuere animalia.
In this book he adds a fifth body which he calls aether, such that along with it the prime bodies are fire, air, water, and earth; and in all of these animals live.
The change from erunt to sint is not palaeographically obvious; that it requires a further change (the insertion of ut) is also worrying. Can ita … erunt not stand (with a full stop or semicolon replacing the comma after uocat)? ‘Thus along with it the prime bodies will be fire, air, water and earth.’ For this use of the future in enumerative summings-up, cf. Chalc. Comm. 288 p. 292.20–1 Waszink non ergo tria sed duo haec erunt initia, … species et silua; 330 p. 325.7–8 erunt igitur tria haec: quod semper est, item quod semper non est, deinde quod non semper est; Boeth. in Porph. comm. sec. 5.12 p. 321.2 Brandt erunt ergo quattuor differentiae quibus genus a differentia, specie, proprio accidentique disiungitur, aliae rursus quattuor quibus … (etc.).
27.23 (p. 124)
de natura loquitur quam putat octo habere potentias secundum quas ratione et ordine mundus administretur. primam autem potentiam solis, secundam lunae, tertiam eorum siderum quae non uocantur erratica; quinque uero <ceteras> illas esse contrarias quae erraticae uocantur.
[H]e … speaks of nature, which he thinks has eight powers according to which the world is governed with reason and order. The first power is that of the sun, the second of the moon, the third of the stars which are not called errant, while the five <remaining> are those which move in the opposite direction, which are called errant.
This passage tracks Epinomis 986a8–b3 fairly closely:
ἴστε ὀκτὼ δυνάμεις τῶν περὶ ὅλον οὐρανὸν γεγονυίας ἀδελφὰς ἀλλήλων … τούτων δ’ εἰσὶ τρεῖς αὗται, μία μὲν ἡλίου, μία δὲ σελήνης, μία δὲ τῶν [πλανητῶν] ἄστρων … πέντε δὲ ἕτεραι.
Since ἄστρων clearly designates the fixed stars, which are distinguished from the five planets, modern editors bracket πλανητῶν. Our author seems to have had it in his text, but he or his Vorlage recognized the problem and tried to make matters right by inserting a negative (otherwise why not simply write fixorum siderum?). Stover's ceteras is presumably motivated by ἕτεραι in the Greek, but I think the transmitted text can probably stand. His contrarias seems right, however. Translate: ‘five in number are those which move’ (etc.).
28.10 (p. 126)
deos autem non putat adfectionibus teneri tanquam uoluptate aut dolore, non magis quam quemquam stultum hominem <…> degere quia solus sapiens ciuilis sit et amicus et notitiam habeat dei.
But he does not think that the gods are constrained by feelings like pleasure or pain, no more than any foolish man lives <happily> since only the wise man is statesmanlike, and a friend, and has knowledge of god.
If these two placita are joined we have to supply putat out of the preceding non putat. I find this difficult, and likewise the use of non magis quam to link two unrelated observations. Does the lacuna instead belong after quemquam? So, for example:
deos autem non putat adfectionibus teneri tanquam uoluptate aut dolore, non magis quam quemquam <sapientem. et negat beatam uitam> stultum hominem degere quia solus sapiens ciuilis sit et amicus et notitiam habeat dei.
The second half of the passage does not seem to fit Ep. 3; amicus might suggest that the author has moved on to Ep. 5.
28.23 (p. 126)
certum denique esse et fidum saluum sapientem.
Finally, the wise man is reliable, trustworthy, and healthy.
Stover restores saluum on the basis of Ep. 10.358c3–4 τὸ γὰρ βέβαιον καὶ πιστὸν καὶ ὑγιές, τοῦτο ἐγώ φημι εἶναι τὴν ἀληθινὴν φιλοσοφίαν, suggesting that the scribe unconsciously substituted a phrase, solum sapientem, that appears several times elsewhere in the treatise (5.3; 19.11; 19.22; 20.4). If so, we may as well write <et> saluum; the corruption would have led to the deletion of the now pointless conjunction.
29.3 (p. 128)
dicit aliquid et de formis quas ἰδέας uocat, incorporales uidelicet substantias quae sint uelut exemplaria quaedam omnium corporum. posse ait tantum aliquid suspicari.
He also says something about the forms which he calls the ideas, that is the incorporeal beings which are, so to speak, certain exemplars of all bodies. He says that it is possible only to suspect something.
Only to suspect something … about what? Surely about the forms. This implies at least a brief lacuna, and in that case we should probably retain possit, for example <et de quibus mens nostra> possit tantum aliquid suspicari.
32.31 (p. 132)
nam materiam sine qualitate sine effigie esse, sed cum a faciente tractatur accipere eam formas et qualitates.
For matter is without quality and without shape, but when it is moulded by the maker it takes on forms and qualities.
The asyndeton bimembre would be isolated in the treatise (on 28.23 see above), and seems excessively ornate for this prosy text.Footnote 5 Contrast Chalcidius, Comm. 310 p. 310.12–13 Waszink siluam sine qualitate esse ac sine figura et sine specie. Should we read <et> sine effigie? Or perhaps sine qualitate siue effigie?
33.1 (p. 134)
in Atlantico (hanc enim <in>scriptionem liber habet) †nullum cuidem† exponit. narrat autem quomodo in Oceano insula quae Atlantis uocatur constitutam habeat rem publicam, et quales sint homines qui in illa morantur.
In the Atlanticus (for the book has this title) he discusses † … †. For he describes how an island in the ocean, which is called Atlantis, has established its government, and of what sort are the people who dwell there.
For the obelized words Stover hesitantly suggests insulam quandam. The use of exponere can be paralleled (cf. TLL 5.2.1762.46–54), but insula in the next sentence looks like the first introduction of the topic. It seems more likely that cuidem is an error for quidem (cf. 3.15 [p. 98], where R has aliqui for alicui), and that there is a contrast between exponit and narrat: ‘he does not expound any <doctrine>, but describes how …’. Compare 9 (p. 106): quomodo dii … bona quidem praestent, numquam autem malorum causas inferant. The missing word might be dogma, which Apuleius has at Plat. 1.4 p. 91.22 Moreschini. Editors there print it as Greek, but the manuscripts give the Latin form. (It is of course also found in the usual title De Platone et eius dogmate, but there is no certainty that this is Apuleian.)
I append a few trivial corrections. At 26.3 (p. 122) read corporis. At p. 121 line 14 ‘but’ should be ‘and’. In the note on 13.2–4 for ‘Rigonis’ read ‘Riginos’ (as also in the bibliography).
