¹ Fritz, Wehrli, Die Schule des Aristoteles, i (Basel and Stuttgart, 1969), p. 231Google Scholar (Critolaus frs. 22–3). In his commentary, pp. 68–9, Wehrli treats these passages as one and the same. He is generally not unaware of the possibility of assigning different versions of the same Greek passage to the same fragment: see, e.g., Critolaus fr. 37a–b, pp. 57–8 and Hieronymus fr. 52a–b, pp. 21–2. Examples could be multiplied. Of the two earlier editions of Critolaus' fragments and testimonia, that of Carpzovius, I. B., Disputatio de vita et placitis Critolai Phaselitae, philosophi peripatetici (Leipzig, 1743)Google Scholar – most probably his doctoral dissertation – has not been available to me. Nor has it been available to the second editor, Frank, Olivier, De Critolao Peripatetico (diss., Berlin, 1895 (see his p. 5, n. 1))Google Scholar, who cites and discusses our passages on pp. 25–6 and 55.

² Cf. Heres 121, πλιν πιστήμης ρχ μν φύσις … πρας δ' ἦλθεν εἰς νθρώπονς Cf. Plato, Phdr. 248a1–b1.

³ As indicated long ago by Carl, Beier (ed.), Tulli, M.Ciceronis De Officiis (Leipzig, 1821), p. 241Google Scholar, on Off. 3.35. Beier's attempt to relate that passage of Cicero to Critolaus has not been followed by more recent editors, rightly. Cicero refers expressly to Aristotle, and is echoing (probably through his source – Panaetius, some doxography, or whatever) passages such as EN 1100bl1–1101a21 (with the metaphor of ῥοπή at 1100b25, and βρθος κα ῥοπή at 1101a29).

⁴ Op. 154,… τν μεγστην τν ρετν θεοσβειαν. Abr.60,…εὐσειας, ρετς τς νωττω κα μεγστης. Sp. Legg. 4.135 …τς γεμονδος τν ρετν εὐσεβεας κα σιτητος. ibid. 147, … τῇ βασιλδι ρετν εὐσεβεᾳ… Decal. 119, εὐσβειαν δ κα σιτητα τν ρετν γεμονσς …

⁵ Plato's list: Gorg. 451el–5; Laws I, 631c; 661a5–6. Three kinds of goods (but with πλοττος or χρματα as the third): Laws 3, 697b2–6; [Epistle 8], 335b. Philo's τρα γνη: see also Mut. 221; Legg. All. 2.20; Post. 112, et al. Philo's δξ as one of τ κτς γαθ: Cher. 117; Det. 136; Post. 112; 117; Gig. 27; Sobr. 61, et al.

⁶ 48, λιγοεῒαν, γκρτειαν, αὐστηρν κα πιστημονικν βον – and esp. 49, τν τς μισονμνης φρονσεως.

⁷ This had already been announced at 42, ζω σὺν αἰσθσει κτ Πρς γνεσιν of 45 is another expression of the same idea. At 48, the five senses are spelled out. At 52, κατ' αἴσθησιν ζω is mentioned again, to return us to our main Biblical lemma. The allegorical interpretation of the two wives, Deut. 21.15–17, is a favourite: see Legg. All. 2.48; Sacr. 19–44, where the beloved wife is δονή (20), offering the pleasures of all five senses (23–4), and the hated wife is ρετή (20), clad in φρνησις (26) and followed by a crowd of virtues (27); Sobr. 21–5, φιλδονος as against φιλρετος (24).

⁸ Nussbaum, Martha C., The Fragility of Goodness (Cambridge, 1986)Google Scholar. Nussbaum's book stops with Aristotle.

⁹ On this simile of the ship, see Anita, Méasson, Du chair ailé de Zeus à l' Arche d' Alliance (Paris, 1986), pp. 176–92Google Scholar (‘Lâme submergée par les flots’). The Platonic antecedents of this simile adduced by her on pp. 176–9 are meagre, and not very convincing.

¹⁰ His chapter ‘The Boat’, vol. 8, 1958, pp. 157–65Google Scholar, is a good example: it treats select passages of Philo, where boats can be interpreted as ‘ psychopomps’ as depicted in Jewish and early Christian art, and neglects our simile of ‘the boat of life’.

¹¹ Tullii, M.Ciceronis Tusculanarum Disputationum Libri V, cum commentario Ioannis Davisii (Oxford, 1805), p. 243Google Scholar. See Davies', edition of De Finibus (λOxford, 1809), p. 336Google Scholar; Klotz', R. edition of Tusc. (Leipzig, 1835), p. 533Google Scholar; Gustav, Tischer's edition of Tusc. (Leipzig, 1850), p. 233Google Scholar. Raphael, Kühner, in his edition of Tusc. (Jena, 1853), p. 428Google Scholar, tries in vain to show that deprimat simply has the required sense. Madvig's, J. N. edition of De Finibus (Copenhagen², 1876), p. 773Google Scholar, and Charles, Anthon's edition of Tusculans (New York, 1852), p. 367Google Scholar – probably independently of each other – assume that Cicero ‘confounded these two distinct metaphors’ (Anthon). But these are not two distinct metaphors: rather, two sides of the same metaphor.

¹² See reference in last note.

¹³ CQ 28 (1934), 127–33Google Scholar, esp. 128–9 (now also in The Classical Papers of A. E. Housman, edd. Diggle, J. and Goodyear, F. R. D. (Cambridge, 1972), pp. 1225–6)Google Scholar. Housman is followed by Gow, A. S. F. (ed.), Theocritus, ii (Cambridge, 1952 (= 1950)), p. 341, n. on 17.195Google Scholar, who merely refers to ‘CQ 28.128’ without author and title. Could it be that at least the Greek evidence had been suggested to Housman by Gow?

¹⁴ Freder, Io.. Gronovii Observationum Libri Quatuor, ed. Platner, Fr. (Leipzig, 1755), pp. 247–8Google Scholar (Lib. II, Cap. XII, fin.)

¹⁵ See Goetz, G. (ed.), Corpus Glossariorum Latinorum, vii (Leipzig, 1901), pp. 551ff.Google Scholar

¹⁶ This would still leave us with the papyrus' περιττα and G's ρἰσδη after ὅλοις. Cohn's interpretation makes the best sense so far. But G's reading may also have come from a gloss ρἰστη κρεἰττονος ζως μοῖρα.