* Preliminary versions of this article were presented at the symposium Crossing Boundaries: Mystical and Philosophical Conceptualizations of the Dunyā/Ākhira Relationship, held on 5 July 2013 at the University of Utrecht, and in a lecture given on 3 December 2013 at the Faculty of Divinity at the University of Cambridge. I am grateful to Christian Lange and Tony Street for these invitations, and to an anonymous reviewer for warning me against the injudicious use of modern philosophical terminology.

¹ Fazlur Rahman, “Dream, imagination, and ʿālam al-mithāl ”, in Gustave E. von Grunebaum and Roger Caillois (eds), The Dream and Human Societies (Berkeley and Los Angeles, 1966), pp. 409–19; Roxanne D. Marcotte, “Suhrawardī's realm of the imaginal”, Ishraq: Islamic Philosophy Yearbook 2 (Moscow, 2011), pp. 68–79.

² Towards the end of §256 of the Ḥikmat al-ishrāq (henceforth: ḤI ), in Shihaboddin Yahya Sohravardi, Œuvres philosophiques et mystiques, vol. 2, ed. by Henry Corbin (Tehran, 1976), pp. 1–260, at p. 241, line 2, al-Suhrawardī uses al-ʿālam al-mithālī. – In what follows, ḤI will mostly be quoted by sections in order to permit consultation of the bilingual edition and translation by John Walbridge and Hossein Ziai (Suhrawardī, The Philosophy of Illumination [Provo, 1999]). Whenever particular lines are cited, reference is to Corbin's edition unless otherwise specified.

³ On Avicenna's eschatology see Herbert A. Davidson, Alfarabi, Avicenna, and Averroes on Intellect: Their Cosmologies, Theories of the Active Intellect, and Theories of Human Intellect (New York and Oxford, 1992), pp. 109–15; Jon McGinnis, Avicenna (New York and Oxford, 2010), pp. 217–21.

⁴ Davidson, Intellect, p. 113. Cf. Ibn Sīnā, Shifāʾ: Ilāhiyyāt, 9.7 (Avicenna, The Metaphysics of The Healing, ed. and trans. by Michael E. Marmura [Provo, 2005], p. 356). Al-Suhrawardī himself restates this theory in his Persian Partaw Nāma (Sohravardī, The Book of Radiance, ed. and trans. by Hossein Ziai [Costa Mesa, 1998], pp. 76f.); cf. also Kitāb al-Talwīḥāt al-lawḥiyya wa-al-ʿarshiyya, in Shihaboddin Yahya Sohravardi, Œuvres philosophiques et mystiques, vol. 1, ed. by Henry Corbin (Tehran, 1976), pp. 1–121, at pp. 89f., lines 15ff., and ḤI, §244.

⁵ Al-Ghazālī, The Incoherence of the Philosophers, ed. and trans. by Michael E. Marmura (Provo, 2000), pp. 208–25.

⁶ ḤI, §§244, 250.

⁷ See n. 1 above.

⁸ ḤI, p. 61, 13f. (§52). – The stipulation shāʾiʿan fīhi bi-al-kulliyya is meant to distinguish accidents from mere parts. Al-Suhrawardī is here trying to improve upon Avicenna's definition of the accident in Shifāʾ: Ilāhiyyāt, 2.1 (Avicenna, Metaphysics, p. 45), part of which he goes on to cite.

⁹ In this respect, al-Suhrawardī's notion of states resembles the modern concept of tropes, on which see Michael J. Loux, Metaphysics: A Contemporary Introduction, 3rd edn (Abingdon, 2006), pp. 71–9.

¹⁰ John Walbridge, “Suhrawardī and Illuminationism”, in Peter Adamson and Richard C. Taylor (eds), The Cambridge Companion to Arabic Philosophy (Cambridge, 2005), pp. 201–23, at pp. 207f. and 210.

¹¹ These two classes of predicates are distinguished in ḤI, §68. On iʿtibārāt ʿaqliyya (a term borrowed from Abū al-Barakāt al-Baghdādī) see ḤI, §§56–68, and my comments in Shihāb al-Dīn al-Suhrawardī, Philosophie der Erleuchtung: Ḥikmat al-ishrāq, trans. with commentary and introduction by Nicolai Sinai (Berlin, 2011), pp. 279–81, p. 332.

¹² See my “Al-Suhrawardī's Philosophy of Illumination and al-Ghazālī ”, forthcoming in Archiv für Geschichte der Philosophie.

¹³ The opposition of the luminous and the non-luminous must not be construed dualistically: a non-luminous substance – a body – is simply a substratum which totally lacks the nature of light. Darkness is thus definable in privative terms: “darkness is simply an expression for the lack of light (ʿadam al-nūr)” (ḤI, pp. 107, 17 – 108, 1 = § 109).

¹⁴ See ḤI, §111.

¹⁵ ḤI, p. 113, 6f. (§117). The formula must not be mistaken for a proper definition, for “nothing is more manifest than light, and nothing is less in need of a definition” (ḤI, p. 106, 13 = §107). – As demonstrated in Sinai, “Al-Suhrawardī's Philosophy of Illumination and al-Ghazālī”, al-Suhrawardī's understanding of luminosity is to a large extent indebted to al-Ghazālī.

¹⁶ The self-consciousness of luminous substances is discussed in ḤI, §§114–120, on which see again Sinai, “Al-Suhrawardī's Philosophy of Illumination and al-Ghazālī”.

¹⁷ Ibn Sīnā, al-Shifāʾ: al-Ṭabī ʿiyyāt, vol. 5: al-Maʿādin wa-al-āthār al-ʿulwiyya, ed. by ʿAbdalḥalīm Muntaṣir et al. (Cairo, 1965), pp. 40–6 (2.2). The passage is also discussed in Paul Lettinck, Aristotle's Meteorology and Its Reception in the Arab World: With an Edition and Translation of Ibn Suwār's Treatise on Meteorological Phenomena and Ibn Bājja's Commentary on the Meteorology (Leiden, 1999), pp. 277–83.

¹⁸ Avicenna mostly appears to use the terms shabaḥ (“appearance” or “image”) and ṣūra (“form”) interchangeably. For present purposes, we may gloss an object's visible “form” as encompassing those aspects of its configuration that are visually perceptible (size, shape, colour, texture).

¹⁹ Ibn Sīnā, Maʿādin, p. 40, 6–10.

²⁰ See McGinnis, Avicenna, pp. 102–10.

²¹ Ibn Sīnā, Maʿādin, p. 40, 12–19.

²² Ibid., pp. 41, 1 – 42, 6.

²³ Ibid., p. 42, 7–18.

²⁴ Avicenna's De Anima (Arabic Text): Being the Psychological Part of Kitāb al-Shifāʾ, ed. Fazlur Rahman (London, 1959), pp. 115–19 (3.5).

²⁵ The same claim is also found in Adelard of Bath (first half of 12^th century), see David C. Lindberg, Theories of Vision from al-Kindi to Kepler (Chicago, 1976), p. 93.

²⁶ Avicenna's De Anima, ed. Rahman, p. 118, 8.

²⁷ Ibid., p. 118, 8–10.

²⁸ Ibid., p. 118, 10ff. (quoting 18f.).

²⁹ Ibid., p. 119, 1f.

³⁰ Ibid., pp. 145, 12 – 147, 14. – The first of the three alternatives sketched by Avicenna would consist in distinguishing between different kinds of inṭibāʿ, yet Avicenna ultimately abandons this approach (ibid., pp. 144, 11 – 145, 11). The third response concedes that a visible object may after all have an impact on the transparent medium located between it and an observer, but opines that this effect may well be imperceptible, i.e., not consist in an imprinting of the object's visual form upon the medium or the mirror (ibid., pp. 147, 15 – 148, 17).

³¹ The text of Walbridge and Ziai has inḍibāḥ, which must be a typo.

³² As ḤI, p. 134, 12f. (§145), will later insist, vision occurs, not by a representation or sensible form of the object of vision somehow coming to be present in the organ of sight, but simply by an illuminated object being opposite a sound eye (bi-muqābalat al-mustanīr li-al-ʿayn al-salīma).

³³ My German translation of §104 (al-Suhrawardī, Philosophie der Erleuchtung, p. 105, 32, and p. 106, 6) twices translates shuʿāʿ as “Lichtstrahl” (“ray of light”), which is capable of being misunderstood as referring to rays of light impinging on the eye rather than proceeding from it (the context clearly shows that reference is to the latter). But note that, as Avicenna remarks in the Najāt, extramissionists “often call the thing which according to them issues from the eye, light” (Fazlur Rahman, Avicenna's Psychology [London, 1952], p. 27).

³⁴ ḤI, p. 102, 5f (§104).

³⁵ Al-Suhrawardī's statement is admittedly elliptical: “the relationship of the crystalline humour of the eye and the objects of vision (al-mubṣarāt) is like the relationship of the mirror” – and what? (ḤI, p. 102, 14f.) Quṭb al-Dīn al-Shīrāzī (Sharḥ Ḥikmat al-ishrāq, ed. by ʿAbdallāh Nūrānī and Mahdī Muḥaqqiq [Tehran, 2005], p. 266, 2) adds “to the forms that are manifest [i.e., in the mirror].” This seems to be the most reasonable guess to me; what becomes manifest in the crystalline humour of the eye in a case of ordinary vision is simply the object of vision (see below), whereas what becomes manifest in a mirror is not the object itself but rather an image thereof.

³⁶ Ibn Sīnā, Dānishnāma-yi ʿAlāʾī: Ṭabī ʿiyyāt, ed. by Muḥammad Mishkāt (Tehran, 1331 AHS / 1371 AHQ), pp. 90, 9 – 91, 1. The comparison between the eye and a mirror is also alluded to, although not explicitly stated, in the Najāt (see Rahman, Avicenna's Psychology, p. 27).

³⁷ On the relationship between the Maqāṣid and the Dānishnāma see Jules Janssens, “Le Dānesh-Nāmeh d'Ibn Sīnā: un texte à revoir?” and “Al-Ġazzālī, and his use of Avicennian texts”, reprinted as chapters VII and XI in id., Ibn Sīnā and his Influence on the Arabic and Latin World (Aldershot, 2006).

³⁸ Al-Ghazālī, Maqāṣid al-falāsifa, ed. by Muḥyī al-Dīn Ṣabrī al-Kurdī, 3 vols. (Cairo, 1936), vol. 3 p. 42, 10–13: the crystalline humour “is like the mirror; if something coloured is opposite to it, then the image of its form is imprinted in it, just as the form of a human who is located opposite a mirror is imprinted in it through the mediation of a transparent body between them”. Al-Ghazālī thus substitutes Avicenna's ontologically non-committal talk of a form in a mirror “shining forth” (tābīdan) from it by the much more fraught concept of inṭibāʿ, the applicability of which to mirror vision Avicenna emphatically rejects in the Shifāʾ. It is true, though, that Avicenna's assertion “If the mirror had a soul, it would see when a form occurs in it” (Dānishnāma: Ṭabī ʿiyyāt, p. 91, 8f.) could be taken to imply that visual forms inhere in a mirror in a manner similar to the eye, as the only difference between a mirror and a percipient subject here appears to consist in the fact that the former lacks a soul. Furthermore, a passage from Avicenna's Epistle on the Rational Soul might also be construed as asserting that forms are properly imprinted upon mirrors: “The [rational] soul becomes as a polished mirror upon which the forms of things become impressed” (cited after Alexander Treiger, Inspired Knowledge in Islamic Thought: Al-Ghazālī's Theory of Mystical Cognition and its Avicennian Foundation [Abingdon, 2012], p. 62). – Incidentally, al-Ghazālī himself (in the Mustaṣfā and Jawāhir al-Qurʾān) denies that mirror images inhere (ḥalla) in the mirror, thus permitting him to maintain that although the human soul can reflect “the divine presence” (al-ḥadra al-ilāhiyya), the latter does not thereby inhere in, or enter into a union with, the soul (Treiger, Inspired Knowledge, pp. 32f.). A similar statement is found in al-Ghazālī's Niche of Lights, where the state of mystical intoxication is compared to someone who looks into a mirror and “supposes that the form he sees is the mirror's form and that it is united (muttaḥida) with it” (al-Ghazālī, The Niche of Lights, trans. David Buchman [Provo, 1998], p. 18).

³⁹ ḤI, pp. 211, 13–212, 3.

⁴⁰ Al-Suhrawardī's use of the root ḥ-l-l may be inspired by the way the Avicennian denial of inṭibāʿ is reformulated by al-Ghazālī (see n. 38 above). That al-Suhrawardī was a close reader of al-Ghazālī can also be shown in other respects (see n. 15 above).

⁴¹ See the qad in the above quotation.

⁴² John Walbridge, The Leaven of the Ancients: Suhrawardī and the Heritage of the Greeks (Albany, 2001), pp. 168f.: “The body, in some mysterious way, is the condition for the form's appearance, but the form is not in the body in the way that the form of the dog is in the body of the dog. Instead the locus makes it possible for the form to be manifest to us – but we see the form, not [only – NS] the locus.”

⁴³ Cf. Edward W. Lane, An Arabic-English Lexicon, 8 vols. (London, 1863), p. 2137b: amruhu muʿallaq = “his affair is left in suspense”.

⁴⁴ ḤI, p. 212, 3f.

⁴⁵ Shams al-Dīn al-Shahrazūrī, Sharḥ Ḥikmat al-ishrāq, ed. by Ḥusayn Ẓiyāʾī (Tehran, 1372), p. 509, 17–20; Quṭb al-Dīn al-Shīrāzī, Sharḥ, p. 450, 8–10.

⁴⁶ Basing himself on the use of muʿallaq and mutaʿalliq in the Arabic Plotinus, Rüdiger Arnzen (Platonische Ideen in der arabischen Philosophie: Texte und Materialien zur Begriffsgeschichte von ṣuwar aflāṭūniyya und muthul aflāṭūniyya [Berlin, 2011], pp. 145–7) construes al-Suhrawardī's use of the term muʿallaq as indicating that some entity is ontologically dependent on, and thus suspended from, a higher-ranking entity. While the parallels adduced by Arnzen are certainly relevant to gauging the intertextual resonance of al-Suhrawardī's terminology, I doubt whether they can settle the question of which precise philosophical meaning al-Suhrawardī himself attaches to the term. Jamal Elias, in an oral response to a preliminary version of this paper, has kindly suggested the possibility that the word muʿallaq might refer to a form's “attachment” to a locus of manifestation, e.g., a mirror. Yet as far as I am aware, al-Suhrawardī nowhere says that suspended images or forms are muʿallaq bi- anything, whether a maẓhar or, as Arnzen would have it, some higher-ranking being.

⁴⁷ Cf. Walbridge, “Suhrawardī and Illuminationism”, p. 209; id., Leaven, p. 162.

⁴⁸ Shīrāzī, Sharḥ, p. 454, 16f.

⁴⁹ Walbridge, Leaven, pp. 168f., fails to make this distinction.

⁵⁰ Must every suspended image have a maẓhar? If one assumes that quasi-substantial images do not pre-exist the moment when they are first beheld or entertained, then it would seem that the answer will have to be affirmative. The “may” (qad) in the above quotation from §225 could be construed as suggesting the possibility of suspended images which do not have a maẓhar; yet it seems equally possible to understand the formulation in the sense of “there is nothing to preclude the possibility of suspended images becoming manifest in maẓāhir”.

⁵¹ It will be recalled that Avicenna, by contrast, assimilates ordinary vision and mirror vision: regardless of whether or not an act of visual perception involves a mirror or not, it is directed at some object in the world.

⁵² John Walbridge, The Science of Mystic Lights (Cambridge, 1992), p. 214 (for the Arabic see ibid., p. 250, 2f.).

⁵³ This is the example used in ḤI, §221.

⁵⁴ The example is taken from al-Shahrazūrī, Sharḥ, p. 506, 2f.

⁵⁵ This does not entail that such self-subsistent images must exist from eternity, waiting for someone to behold or entertain them; they might, for example, come into being when they are first perceived or entertained. This would cohere with ḤI, p. 232, 5f. (§247): “suspended images may newly emerge and be destroyed, like [images] belonging to mirrors and imaginations.”

⁵⁶ Presumably, the substrateless images which we perceive in mirrors, as well as memories, must in some systematic sense correspond to extramental particulars, whereas composite phantasies do not. Yet as far as I can see, al-Suhrawardī does not address this issue.

⁵⁷ ḤI, §§56–60.

⁵⁸ E.g., he proposes to do away with the different classes of propositions and syllogistic figures distinguished in Peripatetic logic (§§21, 25–27), downgrades matter and form from metaphysically real constituents of things to mere mental constructs (§§72–74), reduces the five internal sense to just one (§§222–224), and posits that air and fire do not constitute two different elements but rather the same element at different temperatures (§§195–196).

⁵⁹ See Walbridge and Ziai (ed.), Philosophy of Illumination, p. 99, 16f. (Arabic text). In Corbin's edition, whose text diverges here, the climax ends with two hundred (p. 140, 1). – Janos, Damien, “Moving the orbs: astronomy, physics, and metaphysics, and the problem of celestial motion according to Ibn Sīnā”, Arabic Sciences and Philosophy, 21 (2011): 165–214CrossRefGoogle Scholar, makes it likely that Avicenna was committed to more than just al-Fārābī's ten immaterial intellects, possibly to more than sixty (ibid., p. 200). However, it is entirely possible that al-Suhrawardī could simply have assumed that Avicenna espoused al-Fārābī's position that there are only ten immaterial intellects, for ḤI, p. 155, 4f. (§165) indiscriminately states that “the adherents of the Peripatetics … limit the number of intellects to ten”. And, as §151 of ḤI indicates, even if al-Suhrawardī was aware of Avicenna's position he may nevertheless have felt that the difference between the number of separate intellects in the cosmologies of al-Fārābī and Avicenna was a minor one compared to the plethora of substantial lights advocated by him. In addition, even Avicenna's greater number of immaterial intellects still forms a strictly hierarchical series in which one intellect emanates the following one, whereas al-Suhrawardī also posits immaterial lights which do not form a vertical hierarchy (cf. §§154 and 183).

⁶⁰ ḤI, §§55–60, 63, 89.

⁶¹ On Meinong's proverbial ontological jungle see Dale Jacquette, “On defoliating Meinong's jungle”, Axiomathes 1–2 (1996): 17–42.

⁶² Arisṭū ʿinda al-ʿArab, ed. ʿAbd al-Raḥmān Badawī, 2^nd edn (Kuwait, 1978), p. 72, 8f. I owe my awareness of this passage to Marcotte, “Realm”, p. 76. The comparison of the celestial bodies to an instrument (āla) is found in Shifāʾ: Ilāhiyyāt, 9.7 (Avicenna, Metaphysics, p. 356, 7f.). If the foregoing conjecture is correct, then al-Suhrawardī would have realised that construing the celestial bodies as functioning like a mirror rather than, as suggested by the Shifāʾ, as a “surrogate brain” (Davidson, Intellect, p. 113) held out the promise of conceiving the posthumous experience of heaven and hell in much less subjectivist terms.

⁶³ ḤI, p. 232, 3f.

⁶⁴ ḤI, pp. 229, 10–230, 1. Cf. also id., p. 241, 2 (§256), where it is asserted that someone who was asleep and then wakes up “departs” the ʿālam mithālī.

⁶⁵ Sohravardi, Œuvres, vol. 1, p. 90, 9f.; see Marcotte, “Realm”, p. 76.

⁶⁶ Walbridge, Science, pp. 208f. (see ibid., p. 242, 15, and p. 244, 3f., for the Arabic text).

⁶⁷ “As they desire and will, the wayfarers therein manifest wonders and miracles: the manifestation of their imaginal bodies in various places at one or more times; summoning such food, drink, and clothing as they desire; and the like.” (Walbridge, Science, pp. 208 and 242, 14–17.)