¹ I am very grateful to Ahmad Hasnawi (Paris) for his invaluable comments and suggestions which saved me from many an error. I would also like to thank Lorenz Nigst (Vienna) for a fruitful discussion of the contents of this article and a great number of valuable comments.

The work is described as such on the title page of the Bodleian, MS Marsh 408:

For a description of the work and the conversion of the date see Nasrallah, J., Histoire du mouvement littéraire dans l'église Melchite du V^e au XX^e siècle (Louvain, Paris, 1983)Google Scholar, III.1, pp. 221–2; and also Sepmeijer, F., “The Book of Splendor of the Believer by ʿAbdallāh Ibn al-Faḍl”, Parole de l'Orient, 16 (1990–1991): 115–20Google Scholar.

² MS Marsh 408, fols 1b–2a. Sepmeijer (“Book of Splendor”, p. 115) gives a partial translation of this opening section.

³ See Nasrallah, HMLEM III.1, pp. 221–2; and in greater detail Sepmeijer, “Book of Splendor”, pp. 116–19.

⁴ See Nasrallah, HMLEM III.1, pp. 222–3.

⁵ See Nasrallah, HMLEM III.1, pp. 222–3.

⁶ Sepmeijer, “Book of Splendor”, p. 117, esp. n. 10.

⁷ In the first instance in which al-Anṭākī quotes Aristotle, in Question 4, he refers explicitly to De Caelo, De Generatione et corruptione, and Physica, and in the second instance, in Question 60, to De Anima (on which see below). The mention of Galen in Question 20 is followed by a reference to De Usu partium. See also Sepmeijer, “Book of Splendor”, pp. 116–17; and Daiber, H., “Graeco-Arabica Christiana: The Christian scholar ʿAbd Allāh Ibn al-Faḍl from Antiochia (11^th c. A.D.) as transmitter of Greek works”, in Opwis, F. and Reisman, D. (eds), Islamic Philosophy, Science, Culture, and Religion (Leiden, Boston, 2012), pp. 3–9Google Scholar, pp. 4–5 and esp. nn. 18–20.

⁸ The second half of Question 29 of the K. Bahjat al-muʾmin corresponds to Aristoteles' De Anima. Eine verlorene spätantike Paraphrase in arabischer und persischer Überlieferung, ed. Arnzen, R. (Leiden, New York, Cologne, 1998)Google Scholar, 321.17–323.4; Question 31 to 253.6–15 and 313.7–13; and Question 21 is inspired by 221.7–11. In the second recension of the De Anima-paraphrase discovered and described by M. Sebti (“Une copie inconnue d'une paraphrase anonyme conservée en arabe du De anima d'Aristote: le MS Ayasofia 4156”, in D'Ancona, C. [ed.], The Libraries of the Neoplatonists [Leiden, 2007], pp. 399–414)Google Scholar, the corresponding passages are Istanbul, MS Aya Sofya 4156, fol. 98b3–10; fols 74b7–75a7, but the second parallel is missing; and fol. 69a15–69b8. On Question 31 see also Appendix 2.

In Question 18 al-Anṭākī first paraphrases, then quotes verbatim a passage which corresponds to M. Haji-Athanasiou (ed.), Le traité de Némésius d'Émèse De Natura hominis dans la tradition arabe (Diss. Université Panthéon – Sorbonne, Paris I, 1985), II, 102.10–103.5. See also Appendix 2.

Questions 17, 23 and 24 correspond almost in their entirety to ʿAlī b. Rabban al-Ṭabarī, Firdaws al-ḥikma fī al-ṭibb, ed. al-Ṣiddīqī, M. (Berlin, 1928), 12.11–13Google Scholar; 9.17–10.16 and 10.21–23; and 10.18–20. The part of Question 77 which deals with the opinion of the philosophers on matter quotes verbatim 11.6–13; and in Questions 25 and 28 a couple of sentences from the Firdaws occur, namely 12.19–21 and 11.1–2. On Question 23 see also Appendix 2.

⁹ For the Arabic text of the three questions and answers see Appendix 1.

¹⁰ In Arabic the composite of question and answer is structured as a conditional sentence: “(Question) If someone says … (Answer) then say …”.

¹¹ See Philoponus, Ioannes. De Aeternitate mundi contra Proclum, ed. Rabe, H. (Leipzig, 1899), 58.12–19Google Scholar and pp. 82–4; and the English translation by Share, M., Philoponus. Against Proclus' On the Eternity of the World 1–5 (Ithaca, New York, 2005)Google Scholar of 4.11, p. 51 and pp. 65–7.

¹² I emend the idh of the text to idhā and translate accordingly.

¹³ Al-Anṭākī here refers to Aristotle as ṣāḥib al-manṭiq.

¹⁴ See ed. Rabe, 72.5.

¹⁵ See my edition of the older version with the parallel reading of the Isḥāq's version in the critical apparatus in “The other Arabic version of Proclus' De Aeternitate mundi. The surviving first eight arguments”, Oriens, 40 (2012): 51–95CrossRefGoogle Scholar, pp. 64–8.

¹⁶ I quote the text of D. Gimaret's edition in “Un traité théologique du philosophe musulman Abū Ḥāmid al-Isfizārī (IV^e*–X^e* S.)”, Mélanges de l'Université Saint-Joseph, 50 (1984): 207–52, based on Istanbul, MS Rāġib Pāšā 1463. I have collated it with the readings of Damascus, MS Ẓāhiriyya 4871 which I only indicate in the footnotes in case I have adopted them. I would like to thank Marwan Rashed for lending me his microfiche of the Ẓāhiriyya MS.

¹⁷ I have restricted editorial comments to a minimum, for example not indicating the change to alif maqṣūra. Square brackets give MS readings which I suggest to suppress or alternative readings (iqraʾ … ?).

¹⁸ يلزم ظ : يمنع ج

¹⁹ إن +كان ج

²⁰ The most striking parallels to the older of the two extant Arabic versions of Proclus' De Aet. mundi are (see my edition in “The other Arabic version”, pp. 64–8 with a possible additional emendation suggested based on the comparison with al-Anṭākī's text):

The notion of a possible impediment to the Creator is not found in any of the two Arabic De Aet. mundi versions, but occurs in al-Shahrastānī's account of the first Proclean proof for the eternity of the world in his K. al-Milal wa-al-niḥal, see ʿA. M. al-ʿAbd's edition ([Cairo], 1977) I, 462.14–6:

It seems possible that impediment has indeed been mentioned in the original Proclean argument, namely in the context of the above-quoted discussion of God wanting and being able to make the things resemble Him. The case of His not being able to, which is then mentioned, may well have been phrased including the mention of a possible impediment which could have dropped out in the Greek model of the Arabic De Aet. mundi translations or just have been translated differently in the Arabic Contra Proclum.

²¹ قاله ظ : قال ج

²² في كتابه ظ : ـ ج

²³ إن + كان ج

²⁴ فانا ظ : فانه ج

²⁵ ذلك ظ : ـ ج

²⁶ ان كان يلزم ظ : يلزم إن كان ج

²⁷ قاله ظ : قال ج

²⁸ جواداً ظ : موجوداً ج

²⁹ A reference to God's essence, missing in the K. Bahjat al-muʾmin, also occurs in Shahrastānī's summary of the first Proclean proof (ed. al-ʿAbd, I, 462.11):

إن البارئ تعالى جواد بذاته …

³⁰ فاذن ظ : فيكون ج

³¹ اذ ظ : اذا ج

³² The translation aims at giving an accurate account of the Arabic as it stands. A possible emendation is to delete wa- before majrā and translate: ‘does not cease to be in existence like the existence of the Creator’ which would be similar to the reading of the older Arabic translation of De Aet. mundi (see n. 20 above). Cf. also Isḥāq's translation which reads in its parallel passage ‘the being of the world is like (musāwin) the being of the Creator’.

³³ Al-Isfizārī continues on a far less friendly note with a rebuttal of Philoponus' claim that Proclus has made the world the cause of the existence of God (on which see below). On his criticism of Philoponus here and elsewhere in his K. fī Masāʾil al-umūr al-ilāhiyya see Reisman, D., “An obscure Neoplatonist of the fourth/tenth century and the putative Philoponus source”, in Adamson, P. (ed.), In the Age of al-Farabi: Arabic Philosophy in the 4^th/10^th Century (London, 2008), pp. 239–64Google Scholar, p. 246.

³⁴ The translated reading of the MS Marsh 408 majrāhu jūd al-khāliq does not make much sense here and I have therefore suggested the emendation to majrā wujūd al-khāliq which would translate as ‘like the existence of the Creator’. The emendation seems particularly probable given the similarity in appearance of ه and و.

³⁵ However, one has to bear in mind that the difference between the two words jūd and wujūd in Arabic is only one letter which may easily be dropped or added. When summarising his argument against Philoponus further down, a similar sentence occurs in the K. fī Masāʾil al-umūr al-ilāhiyya, yet reads differently in both MSS. Gimaret (249.1) edits:

whereas MS Zāhiriyya 4871, fol. 144a15 has:

³⁶ Cf. the Aristotelian passage in its Arabic translation (Manṭiq Arisṭū, ed. ʿA. Badawī [al-Kuwayt, Beirut, 1980], 543.6–10):

³⁷ Rashed, M., “The problem of the Composition of the Heavens (529–1610): a new fragment of Philoponus and its readers”, in Adamson, P., Baltussen, H., and Stone, M. W. F. (eds), Philosophy, Science and Exegesis in Greek, Arabic and Latin Commentaries, 2 vols. (London, 2004), vol. II, pp. 35–58Google Scholar, pp. 37–8.

³⁸ In the K. Bahjat al-muʾmin the Greek ‘being diffused in the air’ is rendered as ‘being emanated towards the surroundings of the sphere of [the sun]’ which may be understood as an interpretive translation. The same rendering occurs in the last sentence of this paragraph of al-Anṭākī's quotation.

³⁹ This is hardly a rendering of the Greek. Some correspondence could be achieved by adding – in the same vein as above – ‘wa-al-shams wa-al-ḍiyāʾ maʿan’ or similar, thus making the sun the ‘co-existing’ cause of its light, not a pre-existing one.

⁴⁰ The Greek and the Arabic do not share much similarity except for the notion of darkness. Even if one accepts that this and the previous Arabic paragraph do go back to the Greek, their order is reverse.

⁴¹ This paragraph restructures and rephrases its Greek model. The question concerning the difference between the material and the immaterial light is presented as an affirmation that these two different kinds of light exist, the immaterial of which is said to be perishable. There is also some confusion as to what constitutes material and immaterial light, for al-Anṭākī seem to make sunlight material and immaterial.

⁴² For a translation of the Greek see Share, Philoponus, pp. 26–9.

⁴³ See Contra Proclum, ed. Rabe, 14.21–27.

⁴⁴ Giannakis, E., “The quotations from John Philoponus' De Aeternitate Mundi Contra Proclum in al-Bīrūnī's India”, Zeitschrift für Geschichte der Arabisch-Islamischen Wissenschaften 15 (2002–3): 185–95Google Scholar, esp. 194; Rashed, M., “Nouveaux fragments anti-procliens de Philopon en version arabe et le problème des origines de la théorie de l'‘instauration’ (ḥudūth)”, in Vescovini, G. Federici and Hasnawi, A. (eds), La circolazione dei saperi nel Mediterraneo: filosofia e scienze (secoli IX–XVII), (Firenze, 2013), pp. 323–60Google Scholar, pp. 324–7. It is, of course, also conceivable, but I think rather unlikely, that al-Anṭākī does not quote from a complete Arabic version of Contra Proclum, but a later abridgement. Even more difficult to assess is the relation between al-Anṭākī and the excerpts from Contra Proclum circulating in the guise of Alexander of Aphrodisas, which have been discussed by Hasnawi, A., “Alexandre d'Aphrodise vs Jean Philopon. Notes sur quelques traités d'Alexandre ‘perdus’ en grec, conservés en arabe”, Arabic Sciences and Philosophy, 4 (1994): 53–109CrossRefGoogle Scholar, esp. pp. 68–92. They are often remarkably close to the Greek text, but must have been adapted by an “épitomateur” (see pp. 88–92), whose role in the Greek-Arabic transmission process of Contra Proclum is not entirely clear.

⁴⁵ The number of the book of the De anima referred to has dropped out in MS Marsh 408 but can be supplied with the help of the Greek text.

⁴⁶ This may or may not be an interpretive translation of καὶ γὰρ τούτοις τοῦτο ὑπάρχει ἓν καὶ ταὐτό.

⁴⁷ This must take up τὸ δὲ φῶς οἷον χρῶμά ἐστιν τοῦ διαφανοῦς ὐτό. So the order of al-Anṭākī's account is different from the original one.

⁴⁸ For a translation of the Greek see Share, Philoponus, pp. 30–1.

⁴⁹ The Greek reads ἐνέργεια here and I have no explanation for the Arabic reading.

⁵⁰ Share, Philoponus, p. 30.

⁵¹ Neither the Arabic of the De Anima (418b9–14) as edited by Badawī, ʿA., Arisṭūṭālīs fī al-Nafs, 2^nd edn (Beirut, Kuwait, 1980), pp. 45–6 nor the DA-paraphrase, ed. Arnzen, pp. 257–9Google Scholar or Aya Sofya 4156, fols 75b–76a has any striking resemblance to the present passage. All the texts also have fiʿl, not infiʿāl.

⁵² One could assume that al-Anṭākī had access to some Arabic Aristotle source in which the term infiʿāl instead of fiʿl was used and that he therefore changed the Philoponus text to what he might have considered the more Aristotelian reading. However, the DA-paraphrase, which he obviously quotes in his Questions 29 and 31, has fiʿl as well.

⁵³ I quote the text of the earlier recension of the complete Arabic translation preserved in three Coptic manuscripts which I have reconstructed from the critical apparatus of the edition. On the different translations and recensions see K. Samir, S.J., ‘Les versions arabes de Némésius de Ḥomṣ’, in Pavan, M. and Cozzoli, U. (eds.), L'eredità classica nelle lingue orientali. Acta Encyclopaedica 5 (Rome, 1986), pp. 99–151Google Scholar.

⁵⁴ The change of the examples used here and in what follows is the most striking difference between ʿAlī b. Rabban al-Ṭabarī and al-Anṭākī.

⁵⁵ The text omitted here is the definition of body, which al-Anṭākī quotes in Question 24.

⁵⁶ The text preserved in the MS Aya Sofya 4156 (أص) is closer to al-Anṭākī's quotation and also still unpublished, so I give its reading for the first parallel. The variants of the edited version (أر) can be found in the footnotes. The second parallel is missing in the manuscript, therefore I quote Arnzen's edition. In his edition the first parallel occurs on 253.4–15.

⁵⁷ فلمّا وصف الحواسّ وبيّن الفعل أص: ولما وصف لاحسائس كلّها عامّة ابتدأ في صفة حاسّ حاسّ إلا أنّه قبل أن يفعل ذلك وصف الفرق الذي بين الحواسّ وبين العقل أر

⁵⁸ بين يديها أص : لديها أر

⁵⁹ انفعال أص : فعل أر

⁶⁰ والجزئية أص : والجزئيّات أر

⁶¹ أجساماً اص : الأجسام أر

⁶² كيف أص : لم أر

⁶³ ولذلك اختلفت أص : فلذلك احتاجت أر

⁶⁴ أعني في دركها أر +

⁶⁵ منها أص : ـ أر

⁶⁶ وإلى أن أص : وأن أر

⁶⁷ بين يديها أص : لديها أر

⁶⁸ فإنّما أص : فإنّه ﳌﹽﺎ أر

⁶⁹ فيه لم يحتاج أر +

⁷⁰ بين يديها أص : لديه أر

⁷¹ This probably refers to the above-quoted passage of the DA-paraphrase and may have led al-Anṭākī to combine these two passages.