APPENDIX TO DE PROVIDENTIA
FRAGMENT 1
Really created. In the preceding paragraph, if the Latin translation of the Armenian version is to be trusted, Philo has declared that he is ready to concede “universum ingenitum et sempiternum esse,” a belief which he ascribes not only to Parmenides and Empedocles but also to Zeno and Cleanthes. But still of the “ingenita materia” some part may be created and destroyed (“generetur et corrumpatur”), sometimes by providence, sometimes in the course of nature. He goes on to compare this with the work of a statuary and other craftsmen. According to this hypothesis God did not create eternally the primal matter but used matter to shape the Cosmos. And even if we go a step farther and suppose that the Cosmos itself as well as matter was uncreated (“etsi una cum materia mundus ingenitus supponatur”) there is still room for providence in directing it. In this case the analogy is with the Ephors at Sparta, which they rule though they did not build it. I cannot fit εἰ δὴ γέγονεν ὄντως into this. I should understand it better if for ὄντως we substituted οὕτως = “assuming that this is the method of its genesis.” This is not quite satisfactory, since properly speaking if it is ἀγένητος it has no genesis.
The Armenian has “materiae specialiter factae,” of which Aucher says that the translator read τῆς ὕλης εἶδος. Is it not simpler to suppose that he took εἰ δὴ as a single word and unable to make anything of the rest omitted it?
FRAGMENT 2
§ 4. The thought here is very striking. Wendland cites for it from Sen. Ep. lxvi. 26–27. Here we have “num quis tam iniquam censuram inter suos agit, ut sanum filium quam aegrum magis diligat?… quoniam quidem etiam parentium amor magis in ea, quorum miseretur, inclinat.” But this is not quite the same. For as the sequel “virtus quoque opera sua, quae videt affici et premi, non magis amat, sed parentium bonorum more magis complectitur ac fovet” shows, it is pity for the sufferings of the good and not a yearning for those who have gone astray which Seneca means. Philo’s words come nearer to the spirit of the story of the Prodigal Son than anything I have seen elsewhere in ancient philosophy.
§ 8. περὶ ἃ κηραίνει. This phrase is here given in Gifford’s translation by “about which … are anxious”; in Mangey’s by “quorum in cupiditate … contabescit,” and L. & S. revised, connecting it with κῆρ and citing a very similar passage to this (De Dec. 153), has “be sick at heart or anxious.” But the evidence of Philo’s use of the phrase points to the meaning given in the translation, i.e. “incurring disaster” or “getting into trouble in connexion with something.” Leisegang has eight examples of it, to which add this passage and perhaps De Virt. 31. In none of these is “suffering disaster” impossible and in some “being anxious” is impossible. Thus in Spec. Leg. i. 81 the body of the would-be priest must be scrutinized ἵνα περὶ μηδὲν ἀτύχημα κηραίνῃ; ib. 260 the bodies of the victims sacrificed must be without flaw and the souls of the offerers must κηραίνειν περὶ μηδὲν πάθος; De Praem. 29 the defectiveness of human reason is shown by ὁ λογισμὸς περὶ πολλὰ κηραίνων. In De Ebr. 164 Lot περὶ ταῦτα μάλιστα κηραίνει, where ταῦτα is explained as the fact that Lot had only daughters and therefore could breed nothing masculine or perfect.
§ 17. (Footnote 1, ἄξαντες.) I do not know what sense Dindorf and Gaisford supposed this to have. Gifford, clearly taking it from ἄγνυμι, says that “if it is retained the meaning will be ‘having broken through,’ ” but no such meaning of ἄγνυμι is known, and even if it were possible it would still be necessary to follow it with διά. Nor can any meaning be obtained by taking it from ἄγω. But it is not quite so impossible that it should be the participle of ἀίσσω, though the picture of the physicians being so eager to reach the royal bed that they dart or rush through the bodyguard is, like “breaking through,” somewhat grotesque. In this case we should print ᾄξαντες <διὰ> (though the MSS. would have it without the iota subscript) and ὄχλον and θεραπείαν would be governed by ὑπερβάντες. Wendland suggests as alternatives ἐξ ἐναντίας or ἀντικρὺ or ἀμελήσαντες.
§ 18. σχήματι. Something is to be said for Mangey’s proposal to correct this to ῥεύματι. This is supported by Wendland, but it should be pointed out that in this case the word would be used in the medical sense of a flux or discharge. Galen and Dioscorides both speak of a ῥεῦμα γαστρός or κοιλίας in this sense. The Armenian has a word which Aucher translates by “laxitate” and it is possible that it is some medical term which might indicate discharges or as we should say “looseness” of the bowels, but is διῴδηκε a word which would be joined with ῥεῦμα in this medical sense?
§ 23. (The quotation from Empedocles.) Two lines of this are quoted by Synesius
“ἔνθα φόνος τε κότος τε καὶ ἄλλων ἔθνεα κηρῶν
αὐχμηραί τε νόσοι καὶ σήψιες ἔργα τε ῥευστά.”
Another line quoted by Clement
“κλαῦσά τε καὶ κώκυσο ἰδὼν ἀσυνήθεα χῶρον”
is no doubt rightly supposed to precede the two. The correction of φόνοι τελοῦνται to φόνοι λιμοί τε is apparently due to Stephanus, but I feel as Dindorf evidently did that it is somewhat arbitrary. There is no great similarity between τελοῦνται and λιμοί τε and nothing very strange in Philo quoting the first two words, then inserting the verb, and then quoting the conclusion of the line. Nor is hunger to the point. The places spoken of are those in which not physical evils but human cruelty predominates. The Armenian no doubt had τελοῦνται, for the Latin is “ubi caedes aliaeque huius modi pravae gentium consuetudines vigent.”
§ 24. (Footnote 3, ᾐωρῆσθαι.) This correction of Dindorf for θεωρῆσαι, which is not noticed in Gifford’s later edition, is clearly based on the fact that in Her. iii. 124 Polycrates’ daughter dreamt that her father ἐν ἠέρι μετέωρον ὄντα was washed by Zeus and anointed by the sun. Mangey had suggested μετεωρίζεσθαι. The correction leads up well to κρεμάμενον.
§ 24. The Armenian version of this section as it appears in Aucher’s translation is very curious. Wendland dismisses it as corrupt, but much of it admits of some interesting interpretation. It does not give the name of Polycrates at all, and Aucher in a note says that the translator seems to have read πολὺ κρατεῖ γε, which he rendered by a phrase which Aucher represents by “per multum temporis tenet.” This no doubt he tacked on to the clause about fortune given in the footnote as omitted by Eusebius. He made a full stop then and continued with what Aucher represents by “condigne iis quae patraverat inique impieque ut eorum promotor et auctor sortitus est deterioris vitae infortunium, atque iussu magni regis diu tortus et clavis compressus crudeliter consummatus est.” That is to say he took χορηγός as = “promoter and author” and as subject to ἠδίκησε καὶ ἠσέβησε. At the end of the sentence his “crudeliter consummatus est” seems to represent what he read for χρησμὸν ἐκπιπλάς or perhaps χρησμὸν ἐκπιπλὰς οἶδα. The Latin then proceeds “ilia vero dimiserunt eum quae non multis ante horis gloriae speciem ferebant ante solem ungi et a love lavari.” The words ἔφη κἀμαυτὸν of the received text are to some extent conjectural, for almost all the MSS. divide them otherwise such as ἐφῆκʼ ἐμαυτὸν or ἀμαυτὸν, and if the Armenian by a slight change got ἀφῆκεν αὐτὸν it will explain “dimiserunt eum.” I suspect therefore that he read ἀφῆκεν αὐτὸν τὰ οὐ πρὸ πολλοῦ ἐκτιμῆσαι (or some similar word which he substituted for θεωμῆσαι) δόξαντα, and the translation will run “He was sent out of life by the things which seemed a short time before to have promised him high honour, namely being anointed,” etc. If the similar word is θεῷ εἰκάσαι “to liken him to a god,” we should have something which would make admirable sense and be textually fairly satisfactory, but not well represented by “gloriae.” His version, I am afraid, cannot be accepted in face of the violent changes from the MSS. involved, but it is a much more sensible version. It avoids the pointlessness of putting these words into the mouth of Polycrates and also the contradiction of Herodotus’s story. If we had no access to the Greek and had to choose between his account and that in the translation no one would hesitate to choose the former.
§ 26. ἀνείμονα. For this word see note on De Som. i. 99 (vol. v. p. 599), where this example should have been noted as well as Spec. Leg. i. 83. In all these cases Philo uses this apparently rare word in the sense of without the upper covering and contrasted with γυμνός. The contrast is obvious both here and in Spec. Leg., where it is explained as = “in short tunics,” almost as obvious in De Som. i. 99, where the phrase κοιρᾶσθαι ἀνείμονα means sleeping with inadequate covering. In that note I suggested that Philo had Od. iii. 348 in mind, but if so he misunderstood the meaning, for there the ἀνείμων is not a person who sleeps uncovered but a host who is unable to supply proper covering to himself or his guest. But the misunderstanding is shared by L. & S. which translates it as=“unclad.” I also commented on L. & S. revised being, like Stephanus, still unable to supply an example of the word except that in the Odyssey. In the Addenda however two examples are given, one from a fragment of Callimachus in a papyrus and our Spec. Leg. passage (which however should be given as Ph. 2. 225—not 355).
§ 45. For the Stoic doctrine of “incidental consequences” as distinguished from the “primary works of nature” cf. Gellius vii. 1. 7 “existimat (sc. Chrysippus) non fuisse hoc principale consilium ut faceret homines morbis obnoxios … sed cum multa, inquit, atque magna gigneret pareretque aptissima ac utilissima alia quoque simul agnata sunt incommoda, eaque non per naturam sed per sequelas quasdam necessarias facta dicit quod ipse appellat κατὰ παρακολούθησιν.” This dictum of Chrysippus applies primarily to diseases but the latter part gives it the same general application as Philo gives it here. See Zeller, Stoics and Epicureans, p. 179 (Eng. trans.). Zeller adds that the Stoics also pointed out that things ordinarily regarded as evil may be of the greatest service, and illustrates this from a saying of Chrysippus quoted by Plutarch that bugs do us good service by preventing us from sleeping too long. Cf. for this the incidental uses pointed out by Philo in §§ 47–51.
§ 48. (Footnote 2.) I have allowed what may be called the generally received text to stand but further investigation since the translation was made makes me think that Gaisford and Dindorf were almost certainly right. Gaisford’s App. Crit. seems to indicate that he found τὰ μέτρα or τὰ ἡμέτερα μέτρα in his MSS. with one exception and found πείρᾳ in none. Gilford in the two MSS. which he relied on for this part of the Praeparatio found the same. Also ταῖς ὥραις, not τὰς ὥρας, appears to be universal. On the other hand τῇ ἡμετέρᾳ πείρᾳ goes back to Viger, 1688 and possibly (though I have had no opportunity of verifying it) to Stephanus in 1544. How then did Viger or Stephanus get it? The clue seems to be that the one exception noted by Gaisford has τῇ ἡμετέρᾳ πέτρᾳ. Assuming that Viger or Stephanus found this, the correction to πείρᾳ would be very natural. But if μέτρα is right, ἡρέτερα, which appears in nearly all MSS., must either be dismissed as a dittography or amended to ἡμέρινα (or ταῖς ἡμερίναις … ὥραις?). Wendland, quoting the Armenian, “diei mensuras notat et horas,” suggests τῆς ἡμέρας, but the adjective used in its common antithesis to νυκτέρινος seems to me preferable.
Wendland also notes that the Armenian has “quae de columnis cadunt umbrae,” and suggests that παστάδων should replace ποδῶν.
§ 50. (Quotation from Pindar.) The quotation here alluded to occurs in that part between the two divisions of the second fragment which was omitted by Eusebius. It is undoubtedly from the beginning of a fragment of Pindar preserved in Dionysius of Halicarnassus, De Vi Dem. 6. It is listed among the fragments of Pindar as 107 or 74 (Schröder, p. 427), in Sandys’s Loeb translation, p. 548 as Paean 9. The Latin version in Aucher has enough resemblance to show the identity, but otherwise is sheer nonsense and does not even suggest the general sense, which is that the sun is asked why by this darkening it threatens the world with evil. A version supplied by Conybeare, from which Schröder quotes various bits, would probably explain it better. But it certainly seems that the Armenian who could manage Philo with general accuracy was unable to tackle Pindar. The continuation as given by Dionysius does not suggest the death of kings or the destruction of cities, but war and faction, abnormal storms and floods and through these the destruction of mankind. Some lines however seem to be missed out in the continuation, which may have been more specific.
§ 53. The inconsistency between this and the view expressed in § 41 may perhaps be explained by supposing that though earthquakes, pestilences, etc., are in themselves incidental consequences they may still be employed by God as a means of chastisement.
§ 67. οὗ γῆ ξηρή, κτλ. Zeller in Presoc. Phil. vol. ii. pp. 80–81 (Eng. trans.) has a long discussion on this Heracleitean saying. It is quoted by numerous writers, Stobaeus, Musonius, Plutarch, Galen, Clement and others in various forms and the variation extends to different MSS. of these authors. The chief variants are αὔη ψυχὴ, αὔγη ξηρὴ ψυχὴ, ξηρὴ ψυχὴ. Zeller thinks that αὐγὴ ξηρὴ can hardly be the original form, largely on the ground that there is no such thing as a wet beam. The form οὗ γῆ ξηρή does not appear in any of these quotations, though one variant in the MSS. of Musonius has αὖ γῆ ξηρή, but Zeller has no doubt that this is a true reading in our passage, though his remarks, which are transcribed by Gifford, are oddly worded and not very logical. “Philo,” he says, “ap. Eus. Praep. Evang. viii. 14. 67 has οὗ γῆ ξηρή, κτλ., and that this is the true reading … is clear from the passage in Philo, De Prov. ii. 109 ‘in terra sicca,’ ” etc., i.e. Zeller, unless the translator has misrepresented him, and Gifford certainly, were not aware that Philo ap. Eus. and Philo, De Prov. were the same, and that what he is quoting is only the Latin translation of the Armenian translation of the same passage. What the words in Aucher show beyond doubt is that the Armenian found οὗ γῆ in his text, for he is not likely to have had the acumen to make the correction independently, and they thus give a very convincing support to what we might otherwise have supposed to be an emendation of Stephanus or Viger.
§ 68. (Footnote 1.) The Armenian also presumably read αἰτίου. The full sentence is “mens tamen nusquam nascitur ob frigefactionem gelationemque, quoniam aer, terra et aquae in causis sunt simul, et frequentes exhalationes densae supereminent.” I imagine that he read or translated as if he read ἐξ ἀέρος αἰτίου καὶ γῆς καὶ ὕδατος instead of αἱ γῆς.
§ 71. ἴα. So Wendland from the Armenian “viola vero et rosa crocusque”; this is perhaps the best example of the value which the Armenian occasionally has, see Introd. pp. 449 f. The common reading εἰ does not give any good sense. The rendering which I had given, “though roses, etc., exist they exist for health not pleasure,” lays a difficult stress upon γέγονεν and Gifford’s “roses, etc., are meant, if for health, yet not all for pleasure” misplaces the “if” and gives no clear meaning.