² Burnet, J. (ed.), Platonis Opera, Tomus V (Oxford, 1922)Google Scholar. As Schöpsdau, K., Platon, Nomoi (Gesetze): Buch I–III (Göttingen, 1994)Google Scholar notes ad loc., the conjecture was anticipated (vorweggenommen) by George Burges in a note to his 1852 translation of the Laws. England, E.B. (ed.), The Laws of Plato (Manchester/London, 1921)Google Scholar prints the conjecture and adds in a note ad loc.: ‘I feel sure that Burnet is right in reading λήθῃ for the MS. λήθην.’ des Places, E. (ed. & transl.), Platon. Œuvres complètes. Tome XI: Les Lois, Livres I–VI (Paris, 1951)Google Scholar prints Burnet's conjecture as well; in his apparatus criticus, he notes that the Greek indirect tradition (Galen, Athenaeus and Stobaeus) also transmits λήθην (on the indirect tradition, see id., ‘La tradition indirecte des Lois de Platon [Livres I–VI]’, in Mélanges J. Saunier [Lyon, 1944], 27–40, at 32Google Scholar, repr. in his Études platoniciennes: 1929–1979 [Leiden, 1981], 199–212, at 204Google Scholar). The recent critical edition of Gal. An. mor. corp. temp. confirms that the Greek MSS of Galen unanimously transmit this reading: see A. Μπάζου [Bazou] (ed.), Γαληνοῦ, ὅτι ταῖς τοῦ σώματος κράσεσιν αἱ τῆς ψυχῆς δυνάμεις ἕπονται (Athens, 2011), 69.8. In the Arabic version of this Galenic treatise (Biesterfeldt, H.H. [ed. & transl.], Galens Traktat Daß die Kräfte der Seele den Mischungen des Körpers folgen in arabischer Übersetzung [Wiesbaden, 1973]Google Scholar), Galen's ὥστε [οὖν] ἀνίας καὶ δυσθυμίας λήθην γίγνεσθαι is translated as ḥattā innahū yuṣayyiru l-insāna ilā l-suluwwi ʿani l-ġumūmi wa-ḫubṯi l-nafsi, ‘with the result that [wine] can bring man to forgetfulness of sorrows and (of) ill temper’. Perhaps Galen, who discusses textual variants in his Hippocratic commentaries, had access to both the accusative and the dative readings of λήθη, transmitting one reading in An. mor. corp. temp. and another in his Synopsis; on Galen's approach to textual matters, see Férez, J.A. López, ‘Galeno, lector y crítico de manuscritos’, in Garzya, A. (ed.), Tradizione e ecdotica dei testi medici tardoantichi e bizantini (Naples, 1992), 197–209Google Scholar.

³ The pathways by which the Laws were transmitted into Arabic, and in particular the source used by Fārābī in writing his famous Talḫīṣ (Epitome), have been much disputed. It should first, however, be noted that the ps.-Platonic Kitāb al-Nawāmīs—edited by Badawī, ʿA. in Aflāṭūn fī l-Islām (Tehran, 1974), 197–234Google Scholar—contains no close parallels to authentic Laws material transmitted in Arabic. On this work, see Tamer, G., ‘Politisches Denken in pseudoplatonischen arabischen Schriften’, Mélanges de l'Université Saint-Joseph 75 (2004), 303–35Google Scholar. On another, magical, Kitāb al-Nawāmīs, see Saif, L., ‘The cows and the bees: Arabic sources and parallels for pseudo-Plato's Liber Vaccae (Kitāb al-Nawāmīs)’, Journal of the Warburg and Courtauld Institutes 79 (2016), 1–48Google Scholar.

The first editor of Fārābī’s Talḫīṣ, Francesco Gabrieli, takes no position on Fārābī’s source, simply noting that the bibliographical tradition attributes one translation of the Laws each to Ḥunayn ibn-Isḥāq and Yaḥyā ibn-ʿAdī: see Gabrieli, F. (ed. & transl.), Alfarabius Compendium Legum Platonis (London, 1952)Google Scholar, praefatio, especially ix with n. 2. Leo Strauss's celebrated ‘How Fārābī read Plato's Laws’, in Mélanges Louis Massignon (Damascus, 1957), 3.319–44—reprGoogle Scholar. in What is Political Philosophy? and Other Studies (Chicago, 1959), 134–54—assumesGoogle Scholar that the omissions and apparent misunderstandings in the Talḫīṣ are due to Fārābī’s esotericism. Such an interpretation favours the view that Fārābī may have worked from an integral, or nearly integral, version of the Laws: for a defense of this view, see Mahdi, M., ‘The editio princeps of Fārābī’s Compendium Legum Platonis’ (review of Gabrieli [this note; see above]), JNES 20 (1961), 1–24, at 5–7Google Scholar, and Parens, J., Metaphysics as Rhetoric: Alfarabi's Summary of Plato's “Laws” (Albany, 1995), xxv–xxxivGoogle Scholar. For a philologically grounded analysis of the Arabic evidence, see Gutas, D., ‘Fārābī’s knowledge of Plato's Laws’ (review of Parens [this note; see above]), IJCT 4 (1998), 405–11CrossRefGoogle Scholar, who concludes that Fārābī must have worked from the only version of the Laws that was certainly translated into Arabic: Galen's Synopsis in the version of Ḥunayn ibn-Isḥāq (almost certainly to be identified with the ‘Laws version’ attributed to the same Ḥunayn). More recently, Gutas has summarized the evidence and concluded that ‘[d]’après l'ensemble des témoignages, il semble plus vraisemblable qu'il n'ait existé une traduction arabe que de la seule synopse de Galien et non des Lois en elles-mêmes’: see Gutas, D., ‘Platon – tradition arabe’, in Dictionnaire des philosophes antiques, publié sous la direction de Richard Goulet: Va, de Paccius à Plotin (Paris, 2012), 845–63, at 852–3Google Scholar.

As for the quotations in al-Bīrūnī, Franz Rosenthal suggests that Bīrunī may have used a ‘commented paraphrase’ of the Laws, but Gabrieli concludes that Bīrūnī used ‘una vera e propria versione (s'intende, come al solito, piuttosto libera e qua e là parafrastica)’: see Rosenthal, F., ‘On the knowledge of Plato's philosophy in the Islamic world’, Islamic Culture 14 (1940), 387–422, at 395Google Scholar, with addenda in Islamic Culture 15 (1941), 396–8, repr. in id., Greek Philosophy in the Arab World, art. 2 (Aldershot), and Gabrieli, F., ‘Le citazioni delle Leggi platoniche in al-Bīrūnī’, PP 2 (1947), 309–13, at 313Google Scholar, lightly revised under the same title in al-Bīrūnī Commemoration Volume, A.H. 362 – A.H. 1362 (Calcutta, 1951), 107–10, at 110Google Scholar. Daniel De Smet, for his part, claims that most of Bīrūnī’s quotations ‘sont très confuses ou apparaissent sous forme condensée’, inferring that al-Bīrūnī’s source was a doxography: see his ‘L'Héritage de Platon et de Pythagore: la “voie diffuse” de sa transmission en terre d'Islam’, in Derron, P. (ed.), Entre Orient et Occident: la philosophie et la science gréco-romaines dans le monde arabe (Geneva, 2011), 87–126, at 106–7Google Scholar.

Galen's Synopsis itself, to judge by the extant Timaeus Synopsis, is likely to have preserved enough integral material from the Laws to give the impression, at second-hand, of being ‘una vera e propria versione’, but also to have been paraphrastic enough to license De Smet's judgement. It seems most likely that both Bīrūnī and Fārābī worked from Galen's Synopsis, the former in quoting and the latter in summarizing and reworking. Galen's Synopsis would have almost certainly highlighted material from the Laws about wine drinking, since Galen quotes such passages at length in his An. mor. corp. temp. (see n. 4 below). A shared source would explain, inter alia, an unnoticed parallel, the phrase (man/ʿan) al-sabab fī waḍʿ al-nawāmīs ‘(who was/about) the cause of legislation (lit. ‘of the laying-down of the laws’)’ between the beginning of Fārābī’s Summary and Bīrūnī’s quotation of Leg. 624a. Further, a Galenic source might help to explain Bīrūnī’s quotation of a fourteen-generation pedigree of Hippocrates (tracing his descent back to Zeus), which he states is preserved in an appendix to the Laws: see Sachau, E. (transl.), Alberuni's India – An English Edition (London, 1910), 379Google Scholar.

⁴ MS BNF Arabe 6080 25r.15–18 = MS Köprülü-Fazıl Ahmed Paşa 1001 49v.11–17 = Sachau, E. (ed.), Alberuni's India (London, 1887), 51.15–18Google Scholar = Kitāb fī Tahqīq-ī-Mā li'l-Hind or al-Bīrūnī’s India (Hyderabad, 1958), 71.3–7Google Scholar. Both editions were made on the basis of the Paris MS alone; I have collated both editions and both MSS, all of which transmit the same text of this passage. It should be noted that in An. mor. corp. temp. X, Galen quotes both Leg. 666a3–c2 and Leg. 674a3–c1 on wine-drinking: for the Arabic version of these passages, the first of which is independent from the Bīrūnī quotation under discussion, see Biesterfeldt (n. 2), 35.7–36.1 and 36.10–24, with corrections in id., ‘Ǧālīnūs Quwā n-nafs: Zitiert, adaptiert, korrigiert’, Der Islam 63 (1986), 119–36, at 136Google Scholar.

⁵ See also the translations of Sachau (n. 3 [1910]), 106 (and his corresponding notes at 294) and Gabrieli (n. 3 [1946/1952]), 310–11/108. Gabrieli's translation of part of this passage—‘Dioniso che donò agli uomini il vino come rimedio all'amarezza della vecchiaia, perchè tornassero giovani dimenticando la tristezza, e si mutasse l'animo loro da angustia in salute’—suggests that he interpreted the conjunction wa- as linking a finite verb and a verbal noun within a purpose clause, i.e. (li-) yaʿūdū…wa-ntiqāli = (perchè) tornassero … e si mutasse. The more natural reading of the transmitted text, however, would be (a) to understand the conjunction as linking the two verbal nouns, duhūl and intiqāl, as objects of the preposition bi- ‘by (means of)’. In my transliteration and translation above, I tentatively propose (b) an emendation (of intiqāl to the finite form yantaqila), which would bring the text into closer correspondence with Plato's Greek and, incidentally, Gabrieli's translation. CQ’s anonymous reviewer suggests that (a) may be acceptable, but that for (b) a suffixed pronoun is needed to produce a syntactically sound sentence, i.e. wa-yantaqila ḫulqu nafsihim or anfusihim, ‘and so that the character of their soul(s) might change’. Although the sentence surely reads more smoothly and clearly with the addition of a suffixed pronoun, I do not think that this addition is necessary.

⁶ My emendation posits that the consonantal skeleton of s-l-ʾ-s-h was misread as derived from the much more common root s-l-m (the misreading was almost certainly conditioned by a common meaning of šidda, namely ‘distress, hardship’, the opposite of which would be salāma, ‘safety, well-being’). The emendation is supported by the rendering aslas, ‘more tractable, more flexible’ of μαλακώτερος in the Arabic version of Themistius’ De Anima commentary: see Them. In De An., Gk. (CAG V/3) ed. Heinze, R. (Berlin, 1890)Google Scholar, 1.16 = Ar. ed. M. Lyons (Oxford, 1973), 2.2.

⁷ The Arabic clearly reflects an underlying ἀνηβᾶν at Leg. 666b7, against ἀνίας of the MS tradition of Gal. An. mor. corp. temp. and transmits at Leg. 666b6 τὸν οἶνον, i.e. al-ḫamra, which England (followed by Schöpsdau), Taylor, Müller and Des Places seclude as an intrusive gloss.

⁸ For the ‘amazing correlation’ between the functions of this Arabic particle and those of the Greek dative case, see D. Gutas, ‘Some morphological functions of Arabic Bi-: on the uses of GALex, II’, in Gruendler, B. with Cooperson, M. (ed.), Classical Arabic Humanities in Their Own Terms: Festschrift for Wolfhart Heinrichs (Leiden/Boston, 2008), 62–7, at 65–6Google Scholar, who relies on the data assembled and analyzed in id. and Endress, G. (edd.), GALex, Fascicle 8 (Leiden/Boston, 2007)Google Scholar, s.v. bi- §21.1–8.