Al-Kindī’s Argument for the Finitude of Time in His Critique of Aristotle’s Theory of the Eternity of the World in the Treatise on First Philosophy: The Role of the Perceiving soul and the Relation between Sensation and Intellection

Al-Kindī’s Argument for the Finitude of Time in His Critique of Aristotle’s Theory of the... Abstract The study presents a new interpretation of Abū Yaʿqūb al-Kindī’s (d. ca. 260/873) refutation, in the Treatise on First Philosophy, of Aristotle’s theory of the eternity of the world. Critiquing Herbert Davidson’s classical position that al-Kindī’s three refutations in the Treatise are reformulations of John Philoponus’s (d. 570) in the Contra Aristotelem, the study shows that while al-Kindī’s first and third proofs intersect with Philoponus’s the second one does not. The first part of the study examines the concept of perceptual being (huwiyya) and shows that al-Kindī’s second refutation of Aristotle’s theory of the eternity of the world, which Davidson considered superfluous, is based on a creative reading of Physics 223a16–28. There, Aristotle argues that time is a mind-dependent concept. Because time is dependent on the perceiving soul, al-Kindī distinguishes between time from the mathematical and metaphysical perspectives and time in connection with a body in motion, that is, time from a physical perspective. The second part of the paper, building on the first, offers a fresh perspective on the dichotomous relation between sensation and intellection in al-Kindī’s epistemology. It argues that this problematic relation, described recently by Peter Adamson as the ‘epistemic gap’ between sensation and intellection, could be explained in terms of an epistemic modality. Explaining that the distinction between time mathematically conceived and time physically conceived hinges on a more profound distinction between two epistemic modalities, the study reconstructs the relation between sensation and intellection in al-Kindī’s system. It concludes by indicating the way in which al-Kindī’s epistemic modality anticipates later developments in Islamic philosophy, specifically in the work of Avicenna. Prelude In this paper I offer a fresh interpretation of the second1 of the three counter-arguments Abū Yūsuf al-Kindī (d. ca. 260/873) puts forward in his Treatise on First Philosophy to refute Aristotle’s theory of the eternity of the world. I show that al-Kindī’s critique hinges on the relation his epistemology posits between perception, time and motion. I suggest that al-Kindī’s critique could be read as a creative reconstruction of Aristotle’s argument in Physics 223a16–282 for the connection between time and the soul, and of the reception of that argument by Alexander of Aphrodisias. On this basis, I question the long standing claim that al-Kindī simply reproduced Philoponus’s critiques of Aristotle’s and Proclus’s proofs of the world’s eternity. This claim has been the backbone of the mainstream interpretation of al-Kindī’s critique of Aristotelian metaphysics from the time it was first put forward by Herbert Davidson in 19693 to its recent espousal and reiteration by Peter Adamson.4 Building on the suggested interpretation of al-Kindī’s epistemological commitment to the relation between time, motion and perception, I go on to put forward a new position on al-Kindī’s division of the sciences, especially the distinction between physics and metaphysics. This position enables a new perspective, on the one hand, on the problematic relation between sensation and intellection in al-Kindī’s epistemology, and on the other, the closely connected problem of the relation between empirically particular claims in the domain of physics and universal claims in the domain of metaphysics. The first part of the paper establishes the background of al-Kindī’s critique, starting with the two senses of time presented in Aristotle’s Physics and his conception of the fourfold relationship among time, motion, finite bodies and the perceiving soul. It summarizes Philoponus’s key surviving criticisms in Book VI of the Contra Aristotelem,5 of Aristotle’s definition of motion and its implications for his argument for the eternity of the world. The paper reviews Aristotle’s argument in Physics 223a16–28 for the dependence of time as the measure of motion on the soul as the agent of that measuring. It then discusses Alexander of Aphrodisias’s endorsement of this Aristotelian position in his treatise On Time and its Arabic translation by Ḥunayn ibn Isḥāq (d. 260/873). Alexander’s treatise, as Sorabji and Sharples indicate,6 is one of the key texts in the Aristotelian commentary tradition that discusses in a sophisticated way Aristotle’s thesis on time and the soul. It was also translated around the time of al-Kindī. But I stress that the possible influence of Alexander’s treatise on al-Kindī’s epistemology is incidental to my concern with al-Kindī’s creative reconstruction of Aristotle’s arguments. Alexander’s work is adduced primarily to support the suggested interpretation of al-Kindī’s position on time and the soul. Subsequently, I turn to al-Kindī’s critique of Aristotle’s theory of the eternity of the world in section II of Treatise on First Philosophy. I question Davidson’s conclusion that al-Kindī’s proof was simply a reiteration of Philoponus’s, and Adamson’s commitment to that conclusion despite his indicating (more clearly than Davidson) the differences between the proofs of al-Kindī and Philoponus. I show that al-Kindī’s first and third refutations of Aristotle, which seem to coincide with Philoponus’s are consecutively mathematical and metaphysical, whereas the second is purely physical. I then suggest that al-Kindī’s second proof parallels and goes beyond Alexander’s defence of Aristotle’s argument (found in On Time) for the connection between time and the soul in Physics 223a16–28.7 Thereafter, I show how al-Kindī employs the inseparability of time and the soul to critique Aristotle’s reductio ad infinitum argument in Physics 251a9–28 on the basis of the impossibility of assuming a first or last change. I go on to show that, contrary to Philoponus’s mathematical argument for the finitude of time, al-Kindī’s second refutation of Aristotle’s theory of the eternity of the world is rooted in the view that time is finite inasmuch as it is the measure of motion by a finite perceiving soul. The second part of the paper reinforces the findings of the first part by responding to the challenges implied by the relation al-Kindī establishes between sensation and intellection. It starts with a consideration of the roots of this problematic relation in the opening methodological remarks of the second section of the Treatise on First Philosophy. It then turns to the ramifications of this problem in al-Kindī’s theory of knowledge more broadly, focusing mainly on Adamson’s thesis on the epistemic gap between sensation and intellection in that theory. I argue that al-Kindī’s position on time and the soul is a result of his commitment, expressed at the beginning of the Treatise, to keep the boundaries of the theoretical sciences (mathematics, physics and metaphysics) properly delineated. Al-Kindī holds that the key to this proper demarcation is to take note of the different ways each science corresponds to different modes of perceiving employed by distinct faculties of the soul. This argument supports the relation suggested in Part I between time and the soul in al-Kindī’s epistemology. It also helps connect his conception of time with the intimately related problem of the relation between intellection and sensation. Finally, in Part II of the paper, I try to situate the suggested interpretation of al-Kindī’s epistemology in the context of the broader concerns of early Islamic philosophy, focusing primarily on Avicenna. Part I: The Finitude of Time in al-Kindī’s Critique of Aristotle I.1 The two senses of time in Aristotle’s Physics IV and Alexander’s position In Physics IV.14, Aristotle writes: It is also worth considering how time can be related to the soul; and why time is thought to be in everything, both in earth and in sea and in heaven. It is because it is an attribute or state, of movement (since it is the number of movement) and all these things are movable (for they are all in place), and time and movement are together, both in respect of potentiality and in respect of actuality? Whether if soul did not exist time would exist or not, is a question that may fairly be asked; for if there cannot be someone to count there cannot be anything that can be counted either, so that evidently there cannot be number; for number is either what has been, or what can be, counted. But if nothing but soul, or in soul reason, is qualified to count, it is impossible for there to be time unless there is soul but only that of which time is an attribute, i.e. if movement can exist without soul. The before and after are attributes of the movement, and time is these qua countable.8 In this passage, Aristotle asks the question: if time is the measure of motion should we consider time a property of the substances in motion? But what if there is no one measuring? Would there still be measurement if there were no agent measuring? Would there be enumeration if there were no one enumerating? This simple question provoked much controversy among ancient commentators of Aristotle. Some, like Themistius, rejected the view that time does not have its own real existence (hypostasis). He described Alexander of Aphrodisias as ‘wrong and unfaithful to Aristotle’ for espousing this position.9 Others like Alexander supported this view; he writes in On Time: We say that without a soul to count motions there would be no time [lit: if the soul which counts motions came to nothing, time would come to nothing]. Neither if there were no time would the sphere be in motion. But if it were not in motion, all motions would come to nothing. For it [is the motion of the sphere] which accounts for [lit: is the reason for … ] all of them.10 In this passage, which belongs to the part of On Time where he expounds on Aristotle’s conception of time, Alexander explains that time, as Aristotle suggested, could be considered a quantitative category used by the perceiving soul rather than having an objective reality itself. However, Alexander is also aware that ‘Aristotle himself shows that the before and after in change, which are enumerable, can exist without soul, at least if there can be change without soul … ’11 So he argues that: ‘Perhaps the enumerable will not exist, but that which is contingently capable of being enumerable will exist, such as horses or men, but not the enumerable qua enumerable’.12 In defending Aristotle thus against the possible charge of idealism, Alexander holds that Aristotle’s claim that time depends on the soul does not mean that objects in motion do not exist if there is no soul capable of perceiving them in motion. Rather, Aristotle is saying only that time, that which measures change/motion, will not exist, but the objects in time will continue to exist. Hence, Alexander concludes that ‘ … if there were nothing to enumerate there would be no time. But nothing prevents the substrate of time, which is change, from existing’.13 So, while time as measurement would not exist if there were no agent carrying out the measuring, this does not negate the fact that the object still exists and that it is still in change/motion (kinēsis). Alexander’s defence of Aristotle's argument entails an interesting claim. He interprets Aristotle’s position as rendering time as measurement and not as an objective quality of the substance in change/motion. He implies that time defines the relation between the soul and the objects in motion. So, it is not that perceivable objects depend on the perceiving subject in Berkeley’s sense of esse est percipi. To the contrary, the object in motion will always exist. It is only time, qua the relation between the perceiving soul and the object in motion, that would cease to exist because there is no soul to measure motion and hence there is no relation to explore in the first place. Sharples raises another point related to the core argument of this study. He writes: There is, however, a problem. If time, as the numbered aspect of movement, is dependent on thought for its existence, how can a distinction still be drawn between time in itself as a continuum and time as divided in our thought? It does not seem likely that the explanation is to be sought in a contrast between our intellects, on the one hand, and a superhuman intellect which measures movement without introducing divisions into it, on the other. Rather, as Professor Sorabji has suggested to me, the explanation may be that the instant as it creates time is thought of as travelling, and while being so thought of cannot also be thought of as dividing. Or, putting what is essentially the same contrast in another way, the distinction may be between time as the continuous numerable movement of the heaven and divided time which is movement which is actually numbered.14 Sharples points out the paradox implied by Alexander’s and Aristotle’s position. If time is dependent on the soul, how could Alexander and Aristotle talk about time in itself as a continuous quantity that measures the movement of the heavens? The contrast here is between time as the totality of the actual instants dependent on the presence of a perceiving soul and time as a continuum in which everything changing exists. If time objectively exists as the measure of the continuous movement of the heavens, which have souls and are in motion, how can its existence depend on the finite measuring soul? Sharples suggests, citing Sarobji, that we can speak of continuous time theoretically but if it is not divided then it is not enumerable. So again, without finite souls perceiving similarly finite bodies, there is no enumerable, divided time, the only kind of time that is our concern. Apart from the contrast Sharples draws between Alexander’s position and that of Galen’s to whom he was reportedly responding, the philosophically significant point here is how to reconcile these two positions on time? Further, would such reconciliation still allow for the possibility of talking about time ‘in itself’, the way that Alexander talked about it? I.2 Philoponus’s critique of Aristotle’s theory of the eternity of the world In order to appreciate the subtle relation between al-Kindī’s critique of the eternity of the world and Philoponus’s critique of the same argument, an overview of Philoponous’s arguments is necessary. The surviving fragments from Books I–VI and VIII of Philoponus’s Contra Aristotelem have been pieced together from Simplicius’s commentaries on Aristotle’s Physics and De Caelo in addition to Arabic sources, mostly the works of al-Fārābī (d. 338/950) and al-Sijistānī (d. ca. 390/1000).15 Philoponus devoted Books I–V to his refutation of Aristotle’s arguments for the eternity of the world in the De Caelo and the Meteorology.16 Book I criticizes Aristotle’s theory of the fifth element in De Caelo I.2 claiming that the motion of fire is also circular. While he agrees with Aristotle that motion is determined by the ‘nature’ of the elements, he disagrees that circular motion should be regarded as eternal. This step enables him to question the eternity of the celestial bodies or the sensible non-perishable substances of Aristotle.17 Book II continues with the critique of De Caelo 1.3 269b18–270a12 concerning the weight of the heavens, arguing that the latter do have weight and are consequently natural substances.18 Book III turns to Aristotle’s Meteorology arguing, in line with Plato’s position in the Timaeus, that the heavens are mostly made of fire and hence are not eternal or made from a fifth element.19 In Book IV Philoponus argues that Aristotle’s argument for the eternity of the heavens in De Caelo I.3 270a12–22 is dialectical and not demonstrative.20 Book V mainly argues against the Aristotelian position in De Caelo I.4 that circular motion has no contraries and hence is eternal.21 It is in Book VI that Philoponus turns to the relation between motion, time and sensible substances, directly identifying problems with Aristotle’s conception of that relation. The extant arguments from Book VI can generally be divided into two groups. The first group critique Aristotle’s view that motion is the actualization (energeia) of pre-existing matter. Aristotle argued reductio ad infinitum that matter must be eternal because if matter were generated, then a potentially pre-existing matter must be posited, leading to an infinite regress. This group of arguments constitutes the core of Philoponus’s critique of Aristotle;22 owing to their focus and scope, I will call them ‘physical arguments’. The second group of arguments do not have a central theme; however, they are mostly logical/metaphysical and mathematical. I will thus refer to this group of arguments as ‘speculative and mathematical arguments’. In the fragment from Book VI preserved in Simplicius’s Physics 1129.28–1131.7, Philoponus writes: When Aristotle defined motion as the actuality of the moveable qua moveable, he covered by this definition … all motion in general, yet he assumes that some motions are eternal whereas others possess a beginning and an end. Now, what are his grounds for assuming as a consequence of the definition of motion that things that are going to be moved necessarily pre-exist in time the non-eternal motion, which possesses a beginning of existence, as things that possess only the capacity (dunamis) of motion without the actualization (energeia)?. . Every definition … is predicated of all things defined with equal validity … it follows that from the definition that in the case of non-eternal <motion> the moved object must pre-exist the motion in time, the same will follow in the case of the <eternal motion> too. If this is true the definition is necessarily either (i) not valid … in the case of eternal motion … which is paradoxical. Or (ii) if it were true also in the case of <eternal motion> that the moved object pre-exists the motion in time—so that too the substance of the heavens pre-exists the circular motion in time—<and if> none of the things that possess something pre-existing in time before them is eternal, then eternal motion in this sense will not be eternal … Or (iii) in case they want motion to be eternal, <then> it is not true that it necessarily follows from the definition of motion that the movable pre-exists the motion in time.23 The core of Philoponus’s argument could be paraphrased as follows. Aristotle defines motion as the actualization of the potentially moveable object, implying that the potentially moveable object pre-exists motion. If this definition is correct, then eternal motion would not be possible. For if this assumption were granted, there would be something that precedes the eternal, which is a contradiction a priori. To solve this problem, Philoponus argues: … that when Aristotle gave the definition of motion he did not have the same idea (ennoia) of the potential (to dunamei), <i.e.> as something that can be separated from the actualisation (energeia), that is to say, the motion; but <his idea was> that motion exists as long as the potential is present … and that the potential exists as long as there is motion. For when the motion stops whenever the moved object has reached its completion (telos), then the potential stops as well … Aristotle understood ‘capacity’ (dunamis) as being together with motion and not as <anything> separate from the actualization … 24 In order to overcome the paradoxical implications of Aristotle’s definition of motion, Philoponus re-interprets the relation between potentiality and motion. Both are inseparable. There is nothing that has the potentiality to move and then is actualized through motion. Rather, that which is moveable has the potential to be moved as long as there is motion. The end of motion implies the fulfillment of the potential to move. Philoponus restricts the potentiality and actuality of motion to the interval in which motion/change takes place. This restriction on the concept of motion enables Philoponus to resolve the paradox of eternal motion implied by Aristotle’s definition even though he will refute the possibility of eternal motion later. However, he is interested in pointing out the contradictions entailed by Aristotle’s definition of motion. Philoponus applies the same theory to non-eternal motion. He argues that wood does not possess the potentiality of moving upwards when it is burnt. When wood is burned in fire; this phenomenon amounts to generation and not motion.25 The upward motion of fire is not the actualization of the wood; wood is heavy and naturally moves downwards in contrast with fire which naturally moves upwards.26 It would be absurd to claim that the upward motion of fire resulting from the burning of the wood is the actualization of the latter. Hence, the potentially moveable does not pre-exist the actually moving. Rather, potentiality/capacity (dunamis) is simultaneous with motion/actuality. … motion is the actuality of the immediate capacity. So if fire … and not wood is the potentially movable, and if it is no sooner generated than on the move towards the upper <region>, unless some force prevents it—and similarly the water in the clouds—then it is not true <to say> that in the case of non-eternal motions the potentially movable pre-exists the active motion in time.27 Philoponus manages in this way to undermine Aristotle’s argument for the eternity of motion. For, if no potential thing temporally pre-exists that which is moved/actualized, then the conclusion Aristotle draws that motion and its measure, viz. time, is eternal does not necessarily follow. But an even more important conclusion follows from examining the first set of Philoponus’s arguments against Aristotle’s definition of motion. Philoponus establishes a synchronic inseparability between time, motion and sensible substances. This synchronic relation hinges on a careful consideration of the actuality of motion of sensible substances in experience. It avoids as much as possible any theoretical assumptions. Philoponus bases his physical arguments on the actual physical experience of the different forms of motion that sensible substances exhibit in concreto. This point is in harmony with Frans De Haas’s study of Philoponus’s conception of prime matter as three-dimensional body. As indicated above, Philoponus accepted the principle of concreatio of matter and form, based on the inseparability of form and matter and the impossibility of conceiving incorporeal material substances (that is substances that are material but are not corporeal and hence not three-dimensional).28 Hence, Philoponus argued that if there is any prime matter it is some indeterminate three-dimensional body. This position, which avoids as much as possible theoretical assumptions about the physical world and focuses on the examination of sensible substances as they are actually perceived in finite experience, allows Philoponus to critique Aristotle’s theory of the fifth element. According to Philoponus celestial bodies share obvious characteristics with sensible sublunar substances. Hence, no reasonable ground warrants the assumption that they are composed of a non-perishable fifth element. As De Haas rightly explains, Philoponus wants to confine physics to objects in motion and avoid invoking theoretical speculation about the principles of the science like those about prime matter to explain motion. In so doing, he is more loyal than Aristotle to Aristotle’s own principle found in the Posterior Analytics of not questioning the principles of a science within the science in question.29 Turning to Philoponus’s ‘speculative and mathematical’ set of arguments against Aristotle’s theory of the eternity of the world, I will summarize three of them. The first centres on the critique of the statement, ‘nothing comes to be from nothing’. According to Philoponus this theory logically applies to nature but not to God; otherwise God would not be above nature as required by definition.30 Subsequently, Philoponus refutes three major speculative arguments Aristotle employs to prove the eternity of the world. The first of these is that there must always be time because there is always before and after. In response, Philoponus argues that our tendency as humans to associate what is before with the past and what is after with the future is due to the limitations of the human intellect.31 Angels and God both have the notion of before and after without being in time. Aristotle’s second argument is that almost all physicists are of the same opinion. Philoponus replies that truth is not determined democratically but rather by intellectual enquiry, giving his preference to Plato.32 The third of Aristotle’s arguments is based on the concept of the ‘now’. Since time cannot be conceived without a concept of now and since now is a mean between the past and the future there will always be a time on both sides of now; hence, time is eternal. Philoponus responds by showing that Aristotle commits petitio principii by assuming time in his conclusion. The assumption that ‘now’ is a mean of time is the same as the claim that time is eternal.33 Finally, Philoponus gives three lengthy mathematical arguments against the eternity of the world. The most relevant of these is that if time was eternal then it would be impossible to traverse from an infinite past to a specific point in time. Hence, nothing could be generated from something that is infinite.34 The key characteristic of Philoponus’s second set of arguments against Aristotle’s theory of the eternity of the world is its metaphysical/mathematical nature. On the one hand, Philoponus puts forward a number of speculative arguments like that of the power of God to create ex nihilo by definition. On the other hand, he puts forward a set of mathematical arguments demonstrating that it is mathematically impossible to have anything generated if motion and hence time are infinite. Turning to al-Kindī, I will show that he goes considerably further than Philoponus’s empirical commitment in studying physical phenomenon by examining the epistemic condition of physical experience. I will demonstrate that he expands the role of the perceiving soul not only as the condition of the existence of time, but also as the epistemic basis for determining the different domains of theoretical knowledge and relations among them. This development, as I will argue in the last section, can shed fresh light on the problematic implications of the existential relation between the soul and time that Aristotle and Alexander posited. I.3 Between Philoponus and Alexander of Aphrodisias: al-Kindī’s second proof of the finitude of the world In the second section of On First Philosophy, al-Kindī introduces a number of key methodological distinctions on the boundaries and methods of metaphysics, physics and mathematics. I will return to these methodological remarks in the next section. It suffices here to mention that al-Kindī insists that physical phenomena should not be approached from a mathematical perspective or a logical/metaphysical perspective.35 Subsequently, he foregrounds his proofs of the finitude of the world with a series of definitions and axioms concerning the eternal and the possibility of infinite bodies. Al-Kindī starts, in a manner reminiscent of Philoponus, with a mathematical demonstration that any sensible substance (jirm) must be finite. He follows the same logic in other treatises including ‘A treatise on the essence of what cannot be infinite and what could be called infinite’,36 ‘A treatise to Aḥmad ibn Muḥammad al-Khurāsānī on the finitude of the world body’37 and ‘A treatise to ʿAlī ibn al-Jahm on the unity of God and the finitude of the world body’.38 He then provides a series of logical proofs to demonstrate the impossibility of arguing for an infinite body. For instance, he argues that if there were an infinite body and we imagine separating part of it, the remainder will be either finite or infinite. If the remainder is finite, we must conclude by reductio ad absurdum that no infinite bodies could possibly exist because the sum of two finite quantities must be finite. If the remainder is infinite and if we try to add to it the part we separated, the sum would be the same; the infinite would remain infinite after separation. Since this result does not follow the rules of quantity as applied to possible bodies, then no quantity could be posited as infinite. But this concerns bodies qua quantities, that is from a mathematical perspective. Judging whether or not these bodies can be in finite or infinite motion and hence whether time, qua the measure of motion, is eternal or not, is a question for physics that studies being qua motion, as Aristotle instructs in Book I of the Physics; it is not a question for metaphysics or mathematics. ‘Motion’ is used in the technical Aristotelian sense throughout this paper, the sense it has in Book I of Physics. In discussing Philoponus, I explained the four kinds of motion or change, and the primacy of locomotion. Hence, al-Kindī embarks on a physical inquiry. Following Aristotle in Book IV of the Physics, al-Kindī, approaches the problem of the infinity of time based on a threefold relation between body, time and motion. He starts by asserting that there is no body without time because time is the measure (number) of motion. Motion in turn is not possible except through a body. Al-Kindī writes: If there is a body there is motion otherwise there would not be motion. For motion is some kind of change. For instance the change of the parts of a body or its centre or all its parts is locomotion … there is also growth and diminution … change of qualities is alteration and the change in substance is generation and corruption … change is thus attributed to that which is temporal (dhī al-zamān). Hence if there is motion then there is body … 39 Nonetheless, al-Kindī does not rule out the possibility that time is infinite on a potential level. He does not deny that, mathematically, time is a continuous quantity and hence could be thought of as infinite. However, this is not al-Kindī’s major concern in the context of establishing the relation between time and sensible substances from the perspective of physics; quantities are the object of mathematical inquiry. In this regard, he is mainly concerned with the nature of sensible substances and the legitimacy of the knowledge claims that one can make about them. From this perspective, time, motion and bodies are inextricably connected, each implies the other as far as their possibility in experience is concerned. In order for any sensible substance to be such, it has to be in motion and since time cannot precede motion because the latter is its measure, then the three categories, body, motion and time, all imply each other. Both Davidson and Adamson consider al-Kindī’s position in this passage a reiteration of Philoponus’s physical arguments. However, a careful examination reveals otherwise. As I have shown above, Philoponus’s argument for the inseparability of time, motion and sensible substances is based on his conception of capacity/potentiality (dunamis) as inhering in the moveable object as long as it is moved; the end of motion is equivalent to the actualization of the potentiality. It is true that al-Kindī may have been aware of this Philoponian position from the Contra Aristotelem and the Contra Proclum. But he introduces what Davidson considered a superfluous step ‘ … inferring from the finiteness of time that the ‘body’ of the universe is finite ‘in its being’, and only then, from its finiteness ‘in being’, concluding that the ‘body’ of the universe cannot have existed forever’.40 The term al-Kindī uses for the ‘being’ of the sensible substance is huwiyya. But what does al-Kindī mean by the huwiyya of the sensible substance in this context? The pronoun huwa and the verbal noun (maṣdar), huwiyya, were among the earliest Arabic terms used to denote, respectively, a being and the being of this being. In Kitāb al-Ḥurūf al-Fārābī explains that the pronoun huwa was selected to render the word being/substance because it typically serves as the subject of the sentence, in line with Aristotle’s characterization of substance as the primary subject of predication in the Categories. He explains also that it had been at one point preferred to what became the standard term for being, mawjūd. The reason was the fear of Arab translators of the many implications of mawjūd, a derivative of wajada (to find, etc.). Al-Fārābī, adds that huwa and huwiyya also had a demonstrative/existential sense inasmuch as both indicate to the object perceived.41 In the Epistles of the Brethren of Purity, the term huwiyya is used for perceptual being, as against māhiyya which was used to indicate essence: ‘The perceptual beings (huwiyyāt) of things are acquired by the souls through the senses whereas the essences (māhiyyāt) are acquired by the intellect.’42 Similarly, al-Fārābī in Kitāb al-Fuṣūṣ argues that: ‘the huwiyya of the human being is animal nature and corporeality … ’43 where huwiyya refers to accidents versus the māhiyya—in this case rationality—which refers to the distinctive essence of the human being. Davidson, after textual analysis of of al-Kindī’s second proof, considered it superfluous, due to its complex composition. I will argue that al-Kindī uses the term huwiyya to indicate perceptual being. I mean by perceptual being the being of a being as long as it is perceived by a perceiving soul. Al-Kindī systematically insists that the being of every sensible substance is found within a specific temporal frame of reference (mudda) in which it is what it is. Abū Rīda suggested this meaning (huwiyya) more than half a century ago. In the introduction to his edition of al-Kindī’s treatises he writes: ‘The substantive al-huwiyya is derived from the pronoun al-huwa; the former means the individuated (mutaʿayyin) existence that is grasped by the senses in contrast with the truth or essence of the sensible substance which are grasped by reason’.44Huwiyya and the verb al-Kindī associated with it (yatahawwā) mean to exist to the senses. It denotes the existence (kawn) of the sensible substance in perception; but in order for perception to occur there must be a perceiving soul. Hence, huwiyya denotes the existence of sensible substances in perception that is relative to a perceiving soul. To elucidate this point, I will consider the same passage Abū Rīda uses to articulate his interpretation of huwiyya. In Part II of the Treatise, al-Kindī peculiarly argues: It has been established above that time does not precede motion. Time by necessity does not precede the body (al-jirm); for there is no time without motion [and only bodies are capable of motion]. Hence, no body exists without motion and no motion exists without body and no body exists except within a temporal frame (mudda); this temporal frame is period in which a body is a huwiyya, i.e., the temporal frame is the period in which the body is its huwiyya.45 The gist of al-Kindī’s argument may be reconstructed as follows. Only bodies can be in motion and all bodies are finite. But time is the measure of motion. Hence, time cannot be infinite. In this way, for al-Kindī time, motion and body are inextricably related to each other in the construction of experience. It is untenable, from the perspective of physics or study of being qua motion, to claim that time is infinite since time is the measure of motion and only bodies can be in motion and all bodies are finite. But Aristotle already makes this argument. How does al-Kindī differ from him? The answer to this question hinges on the connection al-Kindī draws between time, motion and bodies through the concept of huwiyya. As explained above, huwiyya refers to the being of a sensible substance in perception (idrāk). Al-Kindī, like early Muʿtazilī thinkers (whose work he knew well), and also in line with their later Ashʿarī rivals, does not reduce the reality (ḥaqīqa) of a thing to its definition (ḥadd) or conception (mafhūm); rather the reality of a thing consists in the definition and conception of it as embodied in its particular, physically perceptible existence. Time measures the period (mudda) in which a thing is what it is (huwa mā huwa)—that is the period in which a material substance is its reality. This reality is essentially the perceptible being of the material substance in concreto. Thus huwiyya is essentially temporal since perception is always in time. This is different from the general Aristotelian theoretical claim that sensible substances and time are connected through motion. Al-Kindī makes a more exacting empirical claim. He is not merely following Aristotle and Philoponus in claiming that there is motion if and only if there are sensible substances. He establishes an immediate connection between bodies and time. He writes, ‘there is no body without a temporal frame’ (lā jirm bi-lā mudda). This statement cannot be interpreted to mean that bodies would not exist if there were no time intervals in which they existed. Rather, it means that bodies are in time because they always are what they really are for a specific interval. This interval has to be in reference to something. I would like to claim that this ‘something’ in reference to which the interval is determined is the perceiving soul. In order to further substantiate this claim, let us consider how al-Kindī’s claim may be better understood in terms of our account of Alexander’s defence of Aristotle’s claims about the dependency of time on the perceiving soul. The core of al-Kindī’s argument concerning the relation between time and sensible substances consists in his insistence on limiting this relation, as Philoponus does, to the boundaries of experience. A sensible substance is always determined as such within a temporal frame, the duration in which it is perceived by a perceiving soul. Perception furnishes access to the physical reality of material/corporeal substances/things. In turn, the reality of anything is the particular, material instantiation of the definition of this thing in its specific individual being; the reality of any human is the definition of the human as instantiated in this particular individual human with all their particular attributes. The perceptual determination of the reality of a sensible substance within a specific temporal frame is denoted by huwiyya. Therefore, the huwiyya of a sensible substance is correlated to the soul being sensible of it; this correlation is determined in terms of time. If this is so, time is constitutive of huwiyya; time cannot merely be an objective measure of motion for the physical reality of the thing is always accessed in time. Hence, al-Kindī gives a rather peculiar definition of time in his Risāla fī ḥudūd al-ashyāʾ wa-rusūmahā ‘Treatise on the Essential/Complete and Inessential/Incomplete Definitions of Things’ (rasm is definition using non-essential differentia, whereas ḥadd is real definition using the essential differentia): ‘Time: is a temporal frame (mudda) measured by motion … ’46 Time does not measure motion; motion measures time. I take this definition to refer to the direct relation al-Kindī wants to draw between time and the perception of the sensible substance. Sensible substances are always perceived in time; time is constitutive of perception. Time cannot be conceived independently of the sense-experience of a specific material substance. Since every sensible substance actually in perception is always in motion and in time in reference to a perceiving soul, there is absolutely no ground for claiming that the world is eternal. Such a claim would be merely speculative and by no means satisfies the conditions of physical investigation. Time defines the relation between the perceiving soul and the sensible substance; such relation is always finite because perception is finite. This argument reinforces, from the perspective of physics, al-Kindī’s previous mathematical proof of the impossibility of an actually existing infinite sensible substance. However, its real value consists in articulating the relation between time and the perception of sensible substances. Against this background it can be argued that the step Davidson considered superfluous is actually rather meaningful, especially as far as it sets al-Kindī’s second proof apart from Philoponus’s physical proofs. Time is finite because it is one of the categories through which a perceiving soul perceives a sensible substance. Based on this and ‘ … on this alone’, as Davidson contends, the sensible substances and their totality, that is the world, must be finite. Sensible substances are always perceived in time and hence in motion and since the former is finite the latter must also be so. But how is this position different from Philoponus’s physical argument? As I indicated in the previous section, Philoponus’s physical argument mainly depends on his adherence to the Aristotelian principle of physics as the study of being qua motion. Avoiding theoretical/speculative assumptions like the positing of pre-existent matter, Philoponus reconstructs Aristotelian concepts of motion and the relation between form and matter in light of actual, observable, finite experience. While, al-Kindī makes a parallel move, his focus on the relation between time, motion and the soul, as against Philoponus’s focus on reconstructing the concept of motion, relocates the critique of Aristotle’s theory of the eternity of the world on different ground. It is true that al-Kindī’s first and third arguments against the eternity of the world parallel Philoponus’s mathematical critiques of the eternity of the world. For instance, in the third argument al-Kindī follows Philoponus’s mathematical argument closely as has been shown by Herbert Davidson, Alfred Ivry and, more recently, Peter Adamson. Adamson calls this argument ‘the counting argument’ and summarizes it as follows. According to Philoponus ‘ … were the world to be eternal ex parte ante then we would be committed to an actual infinity, for example the number of years that have already elapsed. Worse still, we would have an actual infinity that is still increasing. For example, the number of years that had elapsed when Socrates was alive has increased since then, so we would have one actual infinity that is bigger than another, which is (according to both Philoponus and Aristotle) impossible’.47 However, al-Kindī’s argument against the infinity of time that I have explained above, and which Adamson agrees is the longest and most important one in his critique of Aristotle’s argument for the eternity of the world, does not owe much to Philoponus. But if al-Kindī merely echoes Aristotle’s position that time’s existence depends on the soul, a position already expounded by Alexander, what, if any, new contribution did he make to the medieval debate on the eternity of the world? More importantly, does al-Kindī’s position (assuming that my interpretation is correct) help illuminate any further aspects of his philosophy and its contributions to the history of philosophy at large? To respond to the first question, I will turn to the second part of this section. Peter Adamson and Alfred Ivry pointed to a rather paradoxical position that al-Kindī takes in complete opposition to Philoponus. Despite the fact that al-Kindī argued that the world, space and time are finite, he accepted Aristotle’s argument regarding the mathematical/potential possibility of the eternity of the world. He also accepted Aristotle’s argument in the De Caelo for the eternity of the supra-lunar sphere. Ivry writes: ‘[al-Kindī] … accepts the Aristotelian description of the fifth element as simple, ceaselessly moving substance; and agrees with Aristotle’s description of the supra-lunar spheres as not having generation and corruption, being perfectly circular and concentric … Al-Kindī seems to be saying that the world, though not eternal, is in other respects as Aristotle said it was; except it need not be so and would not be so, were it not for God’s will’.48 This paradox conjures up the one Sharples claimed is implied by Alexander’s position in On Time: how can the same thing, time in this case, be deemed as an infinite continuous quantity and at the same time an actual, finite instant? Al-Kindī’s position on the relation between time and the soul affords him ample flexibility in dealing with the two senses of time Aristotle identifies. As Jonathan Lear argued, Aristotle’s account of time operates on two levels: ‘First he gives a theoretical understanding of the nature of time. Second, he wishes to account for our experience of time. These two levels are related.’49 In Aristotle’s Physics VIII.1 251a24–7, Aristotle writes: For if we are to say that, while there are on the one hand things that are moveable, and on the other hand things that are motive, there is a time when there is a first mover and a first moved, and another time when there is no such thing but only something that is at rest, then this thing must previously have been in process of change; for there must have been some cause of its rest, rest being the privation of motion. Therefore before this first change there will be a previous change.50 This passage clearly indicates the contrast between al-Kindī and Aristotle. Aristotle bases his argument for the eternity of world and the infinity of time on the assertion that there are no bodies without motion and that rest is the privation of motion. This is a logical argument based on a theoretical assumption about the concept of motion. Lear comments on this passage as follows: Aristotle is arguing that given any purported first change, there must have been a change, which existed before it. (In a similar vein he argues that given any purported last change, there must be a change which occurs after it.) Thus we can understand his claim that there has always been change as being more than an analytic truth, if we interpret him as claiming that it is absurd for there to have been a first change. Similarly, Aristotle’s claim that the world is eternal should not be interpreted in terms of an infinitely expanded length of time, but only as a claim that no moment could be first or last moment of the world’s existence. Aristotle’s argument establishes no more than the potential infinity of time: time is such that for any moment in time it is possible to find an earlier and a later moment … The fact that infinity of time is only potential is intimately bound up with the role that mind plays. An actual infinite extension of time would, in Aristotle’s eyes, do nothing for us: for we are beings who cannot possibly comprehend an actual infinite extension. But it is the very essence of time to be comprehensible to us. Since the very reality of time is manifested in the soul’s measurements, the infinity of time can be grounded in nothing more than the fact that, given any change, it will always be possible to measure an earlier and a later change … Because our experience of time partially constitutes its reality, Aristotle can infer from our experience of time to the very nature of its existence.51 Lear’s argument in defence of Aristotle’s argument for the infinity of time could be summarized as follows. Aristotle is arguing for the potential infinity of time, which is logically plausible given his assumption that there is no motion without bodies and that bodies are by nature in motion. Rest is a privation of motion; so if a first motion and a first mover are posited infinite regress necessarily follows. Hence, time has to be infinite and the world has to be eternal. However, Aristotle also realizes that we are incapable of comprehending actual infinity; as rational beings we can only posit it potentially. Furthermore, Aristotle recognizes the dependence of time’s existence on the rational soul. Since this is the case we can infer that time is infinite since we are obliged to think of time’s potential infinity. Therefore, it is plausible to infer that time is infinite and that the world is eternal. Al-Kindī’s position has an epistemic advantage over that of Aristotle. If time depends on the soul inasmuch as it perceives a sensible substance, then all assertive claims about time have to be restricted to the finitude of such experience. But in thinking about time on a mathematical level as a continuous quantity abstracted from the world, another set of assertions may be warranted. The problem with Aristotle’s argument, especially in light of Lear’s exposition of it, is that it expands logical and mathematical arguments to the domain of actual, physical experience. The supra-lunar sphere is not sensibly experienced; therefore some speculation can be allowed about its nature. It is in this domain that God’s intervention is allowed and conceived of on a strictly logical and rational basis inasmuch as God is purely transcendent to the world. Al-Kindī radically distinguishes physical arguments from logical or metaphysical ones on the one hand and from mathematical ones on the other hand. For al-Kindī, time, as a continuous quantity, is understood from a mathematical perspective as infinite. This theoretical claim which establishes the ground of the potential infinity of time, as indicated by Lear, must not be conflated with or brought to bear on any judgment concerning the physical nature of the world. The principle of physics is motion, as Aristotle himself systematically asserts in Book I of the Physics. Al-Kindī strictly adheres to this principle. If empirical exploration of sensible substances in motion has revealed to us the essential connection between time, motion and bodies and time has been shown to be connected with the perceiving soul, then the eternity of the world cannot be warranted. It is merely a theoretical claim that has to be distinguished from any assertion about the nature of the physical world. This epistemic distinction affords al-Kindī the possibility of dealing with time differently in accordance with the different principles of each domain of investigation. The difference in principles, as we learn from the Posterior Analytics, respectively determines the approach to beings within each domain. This leads to the second question as to whether this position on time and the soul reflects a deeper commitment in al-Kindī’s philosophy and whether this position makes an interesting contribution to Islamic philosophy at large. Part II: Separating Physics and Metaphysics: The Opening Remarks of Section II of On First Philosophy II.1 Problematizing the relation between sensation and intellection in al-Kindī’s epistemology In the previous section I argued that al-Kindī’s position on time is inextricably connected with his strict division of the theoretical sciences: mathematics, physics and metaphysics. Al-Kindī’s physical argument, as distinct from his other mathematical and metaphysical positions, against the eternity of the world hinges on the epistemological connection between time, motion, corporeality and the soul in concreto. In this section, I further explore the roots of this fundamental division of sciences in the epistemic distinction al-Kindī draws between sensation and intellection. I will show how this epistemic distinction or ‘gap’, as Peter Adamson labels it, indicates the central role al-Kindī assigns to the different and essentially distinct roles of the faculties of the soul in his epistemology. Articulating the centrality of this role has a twofold objective. On the one hand, it reinforces the interpretation argued above of the relation between time and the perceiving soul; on the other, it briefly situates al-Kindī’s epistemological position on the division of the sciences and its roots in the distinct functions of the different faculties of the soul in the broader context of the interests of early Islamic philosophy. I will start with a brief consideration of al-Kindī’s opening remarks in the second chapter of the Treatise and how they fit in a larger context of the problematic relation between sense and intellection and hence physics and metaphysics. Al-Kindī peculiarly opens the second section of the Treatise which he subtitles ‘The first part of First Philosophy’ or the first part of metaphysics with a rather creative reformulation of Aristotle’s Physics 184a17–184b13. Al-Kindī writes: [T]here are two kinds of human perceptions,52 one of which is nearer to us and further from nature. This is the perception of the senses, which belong to us from the beginning of our development, and belong to the genus common to us and to many others, i.e. life, which is common to all animals. Our perceiving with the senses, through the contact of the sense with its sensible object, takes neither time nor effort, and it is unstable, due to the motion and fluctuation of that which we contact, its change in every case being through one of the kinds of motion. Its quantity is differentiated by ‘more’ or ‘less’, ‘equal’ and ‘unequal’, while its quality is contrasted by ‘similar’ and ‘dissimilar’, ‘stronger’ and ‘weaker’. Thus it always occurs in continuous motion and uninterrupted change. It (sc. sensory perception) is that the forms of which are established in the imagination, which conveys them to the memory; and it (sc. the sensible object) is represented and portrayed in the soul of the living being. Though it has no stability in nature, being far from nature and therefore hidden, it is very near to the perceiver, in that his perception is due to the sense, with the contact of the sense with it (sc. the sensible). All sensibles, moreover, are always material, and the sensible is always body and in a body. The other finding/perception is nearer to nature and further from us, being the finding/perception of the intellect.53 In contrast to the same passage in the Physics where Aristotle advises that in any enquiry we should start from ‘the things, which are more knowable and clear to us and proceed towards those which are clearer and more knowable by nature … ’54 al-Kindī advises that we should start from the mode of perception that is closer to us. This mode of perception is immediate sensation that does not require time or effort to discover inasmuch as it is always readily present to us, in contrast to rational perception by the intellect which is more remote from us but closer to nature.55 In other words, al-Kindī stresses that the distinction is between two modes of epistemic comportment associated with two different faculties of the soul rather than laying the emphasis on the distinction between two different objects of knowledge. Having preliminarily distinguished between these modes of human perception, al-Kindī gives a further distinction between them. Sense perception is always representational (mutamaththil). All representations (muthūl) are of sensible qualities (maḥsūsāt); hence, they depend on sense perception.56 By contrast, objects of the intellectual perception are not representational but recognized of logical necessity independently of any representation.57 Representation happens through the internal senses (al-ḥawāss al-bāṭina), specifically through imagination. Imagination in this context is completely reduced to a function related to the re-presentation of substantial material forms that constitute the particularity of every individual substance. Al-Kindī avoids any discussion of the intellectual dimension of imagination in mediating these forms to reason in contrast with his definition of φαντασια in On Definitions and On the Essence of Sleep and Vision.58 Alfred Ivry indicated this point in his commentary on this passage from the Treatise on First Philosophy: It is, in fact, the ‘sensational’ aspect of imagination … , which animals share with men, to which al-Kindī is particularly referring in this sentence, apparently ignoring its more ‘rational’ or ‘deliberative’ character … and it is the latter function which in man prepares his perceptions for comprehension by the intellect. That al-Kindī was familiar with this rationalizing role of the imagination is evident from his definition of tawahhum (ordinarily rendered as ‘estimation’ but used by al-Kindī as the equivalent of ‘imagination’ in the broad sense) in On Definitions … ‘it is … a psychic faculty which apprehends sensory forms in the absence of their matter. It is also said that … imagination is the presenting of the forms of sensible things in the absence of their matter’ … This point is repeated in his treatise On the Essence of Sleep and Vision … Thus it is probably not insignificant that al-Kindī here chooses to refer to a view of sense perception which ostensibly has nothing to do with the intellect; and that he mentions two of the ‘internal senses’ but omits the third and, for man, the most important, viz., the rational faculty, in which perceptions normally culminate. This omission may be due to his desire to draw a sharp contrast between sensory and intellectual perceptions; and to construct a philosophy along lines that are ‘purely’ logical and demonstrative.59 Ivry explains al-Kindī’s interest in focusing on the sensible dimension of the imagination in terms of his ‘ … desire to draw a sharp contrast between sensory and intellectual perceptions’. But he does not explain why al-Kindī is interested in drawing this sharp distinction. Adamson has recently given a rather complex analysis of this contrast. In Al-Kindī, Peter Adamson discusses the contrast between sense perception and rational consciousness at some length. Against the background of his discussion of al-Kindī’s psychology, especially in ‘On the Intellect’,60 Adamson defines two methods of interpreting the complex relation between sense perception and intellection which he respectively describes as the weak thesis and the strong thesis. Adamson describes these two methods as follows: … it will be useful to distinguish between two ways of making intellection, and not sensation, the faculty through which we have knowledge. First, one might hold the relatively weak thesis that sensation and intellection have different objects and that knowledge in the strict sense is only of the objects of intellection, i.e. universals. This would mean that we cannot have knowledge of a particular sensible object as such. But it would not necessarily mean that sensation makes no contribution to knowledge. One might hold, with Aristotle, that sensation somehow makes it possible for us to learn new intelligibles or universals, even though intellection and knowledge grasp universals, not sense-particulars. A second, stronger thesis would be to say that sensation not only has different objects from intellection, but also makes no contribution to intellections. According to this thesis there would be an ‘epistemic gap’ between sensation and intellection, with intellection operating on its own, and sensation serving only to distract us from the good function of intellect.61 Adamson argues that the opening paragraph of the second chapter of the Treatise may support the weak thesis that sensation and intellection have different objects but that sensation may contribute to intellection. However, it does not support the strong thesis that sensation cannot contribute to intellection: ‘If I grasp the universal man … what I am doing is grasping a truth that applies to all particular men. In that case, it seems plausible that we learn the universal precisely through our experiences with various particulars that fall under that universal’.62 Al-Kindī asserts this position in his treatise ‘On the Quantity of Aristotle’s Books’: … the primary substance—I mean the sensible—is known in terms of its predicates—for sensation does not grasp it directly but through the mediation of quantity and quality. Accordingly, the negation of quantity and quality means the negation of the knowledge of primary substance; indeed the true, fixed science of philosophy is the science of substances. Knowledge of secondary substance does not change because its objects do not change and is void of change and its flux. Secondary qualities are initially accessed (yutaṭaraq ilayhā) through knowledge of primary substances.63 However, Adamson argues that sensation is not as central as it may sound pointing to the context of this statement by al-Kindī. Adamson explains that al-Kindī ‘ … is defending the place of the mathematical, propaedeutic sciences as a pre-requisite for all other knowledge. And these mathematical, sciences (arithmetic, geometry, astronomy and harmonics) are not studies of sensible particulars, but of numbers, lines, rations, and the like. Sensation is only required for these sciences insofar as we first encounter number, line, ratio and so on as inhering in a given sensible object’.64 In other words sense perception is only the medium through which we encounter the qualitative and quantitative attributes with which mathematics deals in abstraction from the plenum of the sensible particulars. According to Adamson this does not imply that sensation contributes to knowledge. To substantiate his interpretation Adamson quotes the part of the paragraph that immediately follows: Sensory knowledge is the knowledge of primary substance, and is in flux due to the uninterrupted flux of what is known (this ends only when the [object of knowledge] itself ends, which means that it is wholly destroyed in its substance), or because of the multiplicity of sensible substance and the multiplicity of number. For if all that is numbered is finite, and it is always possible to multiply any number, then what is numbered is potentially infinite in magnitude—if not in the number of the individuals, then in the number of multiples. But what is infinite cannot be comprehended by any knowledge … 65 Adamson takes this passage to reinforce his claim concerning al-Kindī’s commitment to the weak thesis that sensation may only make possible the knowledge of new universals. Subsequently, Adamson reverts to al-Kindī’s argument in the opening passage of the second chapter of the Treatise regarding sensible particulars. All the sensible particulars are in a state of flux with regard to quantity and quality, which, in turn, are the basis of any possible form of knowledge, as indicated above. Adamson adduces al-Kindī’s claim that sensible particulars can comprise opposites; for instance a sensible particular can be both equal and unequal at the same time (for instance a stick can be equal in length to one stick and not equal in length to another). He suggests that al-Kindī’s position be read alongside Plato’s similar claim in the Phaedo. Hence, on the relation between sensation and rational cognition, he writes: ‘Sensation merely prompts the soul to uncover the intelligibles that are already within it, by reminding it of those intelligibles. The intelligibles cannot be somehow gleaned afresh from sensation’.66 Adamson uses al-Kindī’s discussion of the knowledge as recollection of universal intelligibles apart from any material differentiation in his treatise ‘On Recollection’ to suggest that al-Kindī may have very well endorsed the strong thesis, viz. that sensation does not have any epistemological function. Thus, Adamson argues that, in a number of his writings, al-Kindī leaves an ‘epistemic gap’ between sensation and intellection indicating that ‘he seems unaware or unconvinced by Aristotle’s strong empiricist commitments … ’67 In other words al-Kindī was not convinced that universals could be abstracted from sensible particulars. Adamson cites a passage from al-Kindī’s treatise On Rays, which challenges his claims about the Platonism of al-Kindī. In this passage, al-Kindī suggests that reason uses sensation to grasp the common forms, the universals shared by sensible particulars. Adamson concedes that his Platonic conception of the ‘epistemic gap’ between sensation and intellection in al-Kindī’s epistemology could be questioned. However, he acknowledges this only as ‘the exception that proves the rule’ and on the assumption that On Rays ‘was written late in al-Kindī’s career’.68 However, the sharp contrast al-Kindī draws between sensation and intellection has complex methodological and epistemological connotations that extend beyond the challenge to Adamson’s Platonic description of its origins. Methodologically, this contrast follows al-Kindī’s assertions in the first section of the Treatise that the different domains of science that inquire into different regions of beings, in turn accessed through different faculties, have to be separated. Hence, al-Kindī advises that inquiries into non-sensible substances must employ purely rational arguments; the science that encompasses all such inquiries is metaphysics. Similarly, mathematics must not be used in physics proper since physics is the study of sensible substances qua motion or change.69 Different regions of beings that are accessed by different faculties are to be approached through different domains of science. Epistemologically, the ramifications of such a contrast are rather ambiguous. As shown above, Adamson demonstrated that al-Kindī’s characterization of the relation between sensation and intellection is not always consistent. However, there is an important strand to this complex picture worth noting. While it is indisputable that al-Kindī would consider knowledge proper to be pure thinking of universals devoid of any sensible qualities, al-Kindī seems to be consistent in asserting that both sensation and intellection are modes of finding out that complement each other epistemologically. In his treatise titled ‘On Dreams’, al-Kindī writes ‘ … therefore Plato said that the soul is the receptacle of all things sensible and intelligible. Hence, the soul knows by nature. For all knowledge comes from reason or the senses … ’.70 This sheds new light on the relation between sensation and intellection. Beyond the weak and strong theses suggested by Adamson to define the relation between sensation and intellection, it seems that al-Kindī is more interested in identifying two types of perception that equally contribute to our knowledge of the world even if knowledge proper is that of rational universals. In other words, al-Kindī is interested in defining the two domains of perception as two modes corresponding to two faculties through which the soul can know or at least relate to a knowable thing. The same thing can be approached differently from the respective perspectives of these two domains. This difference is not due to a change in the essence of this same thing; rather, it depends on the mode of finding out applied to it, which in turn determines what of its characteristics it will disclose. The interpretation of al-Kindī’s argument for the finitude of time suggested in the first part of this study supports this interpretation of the contrast between intellection and sensation. Shifting the focus of concern from the object to the soul that perceives the object enables al-Kindī to respond to the problematic implications of the thesis that the existence of time depends on the soul. Time qua intellectual perception can be thought of mathematically as a continuous quantity. Further, metaphysically speculative claims may be made about time as such in connection with essentially rational concepts/problems, i.e., concepts/problems that are not part of nature. Time qua sensation is actual, divided time that is inseparably connected with motion and sensible substances, the immediate object of sensation. This is the advantage that al-Kindī’s position has over that of Philoponus on the one hand, and Alexander and Aristotle, on the other. The former critiqued the theory of the eternity of the world by reconstructing the concept of motion, restricting himself to, essentially, physical parameters and avoiding all theoretical speculation. Al-Kindī develops this position by showing the essential role of the soul as a necessary condition for physical inquiry. While Alexander and Aristotle pointed out the constitutive role of the soul, al-Kindī’s insistence on separating the different domains of theoretical sciences as corresponding to different modes through which the soul comports itself towards the same thing enables him to overcome the difficulties entailed in this position. To reinforce my interpretation, I will make one last suggestion about the epistemic nature of the moral distinction between perceptual being and rational/ideal being in al-Kindī’s system. II.2 Between al-Kindī and Avicenna: the logical vs. epistemic modality of being and the relation between sensation and intellection Adamson has argued that al-Kindī uses two distinct yet reconcilable notions of being.71 The first is simple being. Simple being is the being of God to whom unity is exclusively attributed. In this vein, simple being is being devoid of any essence or attributes. Hence, simple being in itself does not allow for predication. Simplicity, here, consists in that being cannot be qualified. In this sense simple being is different from unlimited being. Adamson explains that this conception of simple being is distinct from the concept of unlimited being endorsed by the Muʿtazili school of rational theology. According to this school, God is unlimited because He is identical with His attributes/predicates. In contrast, Adamson argues that, for al-Kindī, God is simple being in itself because nothing can be predicated of Him. Therefore Adamson likens al-Kindī’s sense of simple being to Aristotle’s sense of prime matter: ‘ … being’ [that is simple being] is treated as analogous to Aristotelian matter … [Aristotle] … describes matter as the ultimate subject of predications that underlies all features of a thing. Also like Aristotelian matter, being subsists through change as becomes clear … when al-Kindī says that being ‘does not change’. The point is an intelligible one: even in the case of substantial corruption (such as death in the case of a human), there is not an absolute destruction of being but merely of the way the thing is. This is why the corpse that remains when the human is no longer alive is yet something that exists. Finally, like Aristotelian matter, mere being must be simple, where ‘simple’ again means without predicates. For as the ultimate subject of predication, being itself cannot be further analyzed into a complex of subject and predicates.72 By contrast to simple being, complex being is the being of the world, everything that is generated. Adamson paraphrases al-Kindī on complex being as follows: ‘ … the body of the universe causes it to have a limited, temporal existence. Here it is more natural to understand anniyya as “nature” or “essence”. Indeed, at one point he makes a remark that equates huwiyya, “being” with mā huwa “what a thing is” ’.73 Adamson summarizes al-Kindī’s general position on being as follows: When he [al-Kindī] speaks of ‘being alone’, he means the mere act of existing that is prior to, and the subject of, the existent’s essence and other predicates. But he also speaks of ‘a being’, by which he means a fully constituted being that is already considered to have an essence. On this latter notion, the being of each thing will be distinct from the being of anything else; on the former notion, being is mere existence and belongs to anything that God has seen fit to create.74 Adamson argues that al-Kindī’s concept of simple being anticipates some aspects of Avicenna’s metaphysics associated with the distinction between essence and existence. Both Avicenna and al-Kindī maintained that being is distinct from attributes, whether essential or not. For instance, existence for Avicenna can be thought in complete independence of essence. Similarly, God for both Avicenna and al-Kindī is pure existence, which, in turn, instantiates the complex existence of the essence of every other being. Adamson contends, however, that al-Kindī’s concept of complex being does not show any aspects that could possibly anticipate Avicenna’s distinction between essence and existence. For al-Kindī the existence of object X consists in satisfying the essential features of X. If X is a human being, the essential features would be animality and rationality. Similarly, al-Kindī and Avicenna cannot be reconciled on the priority of essence over existence: al-Kindī is committed to the priority of existence—the reverse of Avicenna’s position.75 I would like to suggest another parallel between al-Kindī and Avicenna, specifically concerning the relationship between the essence and existence of complex being. Olga Lizzini has argued that what is most distinctive about Avicenna’s metaphysics is its modal orientation.76 She suggests that what is most provocative about Avicenna’s notion of being is not whether or not essence precedes existence, as many scholars have tended to remark. Rather, according to Lizzini, what matters most is whether the existent under consideration is necessary or possible, as Avicenna argues in book I.6 of the Ilāhiyyāt of his al-Shifāʾ. In other words, Avicenna gives priority to the logical determination of the nature of the essence and whether or not it is necessary rather than possible over and above whether or not it is existentially instantiated. A similar modal orientation can be discerned in al-Kindī’s epistemology. He identifies two modes of finding out: intellection and sensation. These two modes identify two modalities through which any substance can be epistemically accessed. In turn, these two modalities allow for different senses of being of the same substance. As argued above, the being of time mathematically considered can be potentially infinite. However, its being is finite in actuality. On a mathematical level it is a continuous quantity that is infinitely divisible. In actuality, it is associated with the ‘being’ (huwiyya) of a particular sensible substance as grasped by a perceiving soul. Since any sensible substance is constantly in motion and since time is the quantity measuring this motion, then time must be finite. This is another way of thinking about the relation between simple and complex being. Simple being in relation to every substance defines the modality in which it is conceived and presented apart from its essence and attributes. Hence, it is epistemic modality that determines the nature of being of the essence, and not vice versa. Unlike Avicenna’s strictly logical conception of the modalities of essences, I am suggesting that al-Kindī associates modality with the faculties of the soul. Following the Aristotelian line of thinking in the De Anima, al-Kindī associates each faculty of the soul with a specific domain of beings. However, in contrast with Aristotle (and definitely more lucidly), al-Kindī argues that the being of certain classes of beings can be understood differently based on the way they are considered by different faculties of the soul. In addition, there is a further connection between the reception by different faculties of the soul of the same object and the constitution of the domains of sciences in al-Kindī’s epistemology. In the domain of physics senses and empirical observation play a definitive role in the constitution of judgments about nature. By contrast, the domain of metaphysics essentially relies on logical investigations of universals and relations among them just like the domain of mathematics, in its own right, relies on abstract quantitative (kammī) analysis. The results of the investigations in each scientific domain cannot be extended to a different domain because their principles are different. Notwithstanding the distinction I have just expounded, between al-Kindī’s epistemic modality and Avicenna’s insistence on the constitutive role of logical modality, I would like to argue that both are inspired by a similar critique of the primacy of metaphysics in classical antiquity. Let me begin with this important observation by Charles Kahn in ‘Being and Existence’: … My general view of the historical development is that existence in the modern sense becomes a central concept in philosophy only in the period when Greek ontology is radically revised in the light of a metaphysics of creation; that is to say under the influence of biblical religions. As far as I can see, this development did not take place with Augustine or with the Greek Church Fathers, who remained under the sway of classical ontology. The new metaphysics seems to have taken shape in Islamic philosophy in the form of a radical distinction between necessary and contingent existence: between the existence of God on the one hand and that of the created world on the other. The old Platonic contrast between Being and Becoming, between the eternal and the perishable (in Aristotelian terms between the necessary and the contingent) now gets reformulated in such a way that for the contingent being of the created world (which was originally present only as a possibility in the divine mind) the property real existence emerges as a new attribute or ‘accident,’ a kind of added benefit bestowed by God upon possible beings in the act of creation. What is new is the notion of radical contingency: not simply the old Aristotelian idea that many things might be other than they in fact are—that many events might turn out otherwise—but that the whole world of nature might not have been created at all: that it might not have existed. I leave it to the historians of Islamic and medieval philosophy to decide how far my hypothesis is correct and to determine just when, or in what stages, the new concept of existence was formulated … 77 With that observation in mind, it will be worthwhile to indicate how my interpretation of the epistemic gap between sensation and intellection in al-Kindī’s theory of knowledge could clarify how his critique of Aristotle’s theory of the eternity of the world contributed to the reconceptualization of the question of being in Islamic philosophy—no less than did Avicenna’s logical distinction between necessary and possible essences did. Al-Kindī does not definitely relegate the problem of existence in the way that Avicenna does through his distinction between essence and existence and his emphasis on the fundamentality of the logical modality of essence. However, al-Kindi’s distinction between sensation and intellection as two possible epistemic modes enables a reconceptualization of the relation between epistemology and metaphysics in a manner that was inconceivable by Greek ontology. If the interpretation of the epistemic gap suggested above is correct, then al-Kindī’s position on complex being can be revisited as follows. Just like ‘time’, the being of any complex being depends on the epistemic mode through which it is being considered. From a purely rational/metaphysical perspective that being would be defined in terms of the existence of its essential attributes; from a physical perspective it would be the huwiyya or how these essential attributes are constituted for a perceiving soul. In this regard, al-Kindī may be anticipating the later distinction in Islamic philosophy and philosophical theology between mental existence (al-wujūd al-dhihnī) and perceptual external existence (al-wujūd al-khārijī). This was later formalized in the twelfth century under the influence of al-Ghazālī (d. 505/1111) and Fakhr al-Dīn al-Rāzī (d. 606/1209). What is more certain is that al-Kindī’s epistemic gap reflects or at least makes possible a reformulation of the ontological commitment of Greek naturalism. Embedded in this reformulation is a favouring of epistemic conditions over the concern with metaphysical claims. If Charles Kahn’s interpretation of the significance of Avicenna’s distinction between essence and existence and the necessary and the contingent in terms of re-questioning Greek ontology are correct, I would argue that al-Kindī epistemic modality contributes to the same objective. Conclusion In this paper I have pursued two objectives. The first was to clarify the importance of al-Kindī’s peculiarly formulated second counter proof of the eternity of the world. Contrary to Davidson’s thesis, and its adoption and reformation by Adamson, I showed that this step is not superfluous. Rather, it leads to an important aspect of al-Kindī’s thought, which reconstructs and creatively develops, a classical Aristotelian position that was espoused in the tradition of Aristotelian commentators by Alexander in On Time. The conclusion drawn about al-Kindī’s endorsement of the relation between time and the soul indicates that, while al-Kindī was influenced by Philoponus’s critiques of Aristotle, he sought to find a new ground for the latter’s interest in restricting physical arguments to actual, finite experience. My second objective was to reinforce this suggested reading of the second proof by indicating its connection with al-Kindī’s commitment to the demarcation of the boundaries of theoretical sciences. I argued that this demarcation hinges on the correspondence between these sciences and the two different modes through which the soul comports to objects: sensation and intellection. If this interpretation is correct, this helps mitigate key problematic implications, as Sharples indicated, of Alexander’s and Aristotle’s position on the dependence of time on the soul. Further, the findings of the second section sought to clarify the ‘epistemic gap’ between intellection and sensation pointed out by Alfred Ivry and further developed by Adamson. I have tried to situate this argument in the broader context of al-Kindī’s epistemological commitments, explaining how it is in harmony with the findings of the first part of the paper, and exploring briefly some affinities it shares with Avicenna’s modal interests. Footnotes 1 Al-Kindī advanced three major counter-arguments to refute Aristotle’s and the Neo-Platonist’s position on the eternity of the world. In this paper I focus on the second one. The order of the arguments I follow is that of Herbert Davidson in his landmark article: ‘John Philoponus as a Source of Medieval Islamic and Jewish Proofs of Creation’, Journal of the American Oriental Society, 89/2 (1969): 357–91. 2 All the translations of Aristotle’s works cited in this paper are quoted from Jonathan Barnes, The Complete Works of Aristotle (Princeton, NJ: Princeton University Press, 1995), with references in accordance with the Bekker/Prussian Academy of Sciences standard pagination. The Physics translation is by R. P. Hardie and R. K. Gaye. 3 Herbert Davidson, ‘John Philoponus as a Source’, 357–91, at 370–5. 4 Peter Adamson, Al-Kindī (New York: Oxford University Press, 2007), 74–105. See also Peter Adamson and Peter E. Pormann, The Philosophical Works of Al-Kindī (Karachi: Oxford University Press, 2012), 3–81. 5 Brief references will also be made to Philoponus’s other major critique, in Contra Proclum, of Proclus’s appropriation—in On the Eternity of the World and his lengthy commentary on the Timaeus—of Aristotle’s argument for the eternity of the world. 6 See section I.1 below for references and some discussion of the work of Sorabji and Sharples. 7 While the existential connection between time and the soul was acknowledged in the works of a number of ancient commentators including Simplicius, I take note of Alexander of Aphrodisias’s treatise On Time because of its importance and its having been translated around the time of al-Kindī. The treatise was translated by Ḥunayn as is stated on the Arabic manuscript discovered and published by ʿAbd al-Raḥmān Badawī in 1971 (see n.10 below). It was mentioned in Abū al-Rayḥān al-Bīrūnī (d. 439/1048), Kitāb al-Bīrūnī fī taḥqīq mā li-l-Hind min maqūla maqbūla fī al-ʿaql aw mardhūla (Hyderabad: Dāʾirat al-Maʿārif al-ʿUthmāniyya, revised edn., 1958), 271–6. It is difficult to determine historically when Ḥunayn ibn Isḥāq did the translation and whether it could have been available to al-Kindī when he most likely composed his Treatise during the reign of the Caliph al-Muʿtaṣim (r. 218–27/833–42). However, given the overlap between Ḥunayn’s and al-Kindī’s time it is not unreasonable to venture that the latter was somehow familiar with it. Nonetheless, Alexander’s position on the nature of time is invoked here as a philosophical forerunner to the position that, as this study suggests, al-Kindī takes in his second proof. The reference to Alexander’s On Time is not intended as a historical claim that al-Kindī actually read it, and the validity of the argument here is not dependent on the validity of such a claim. 8 See R. P. Hardie and R. K. Gaye, The Complete Works of Aristotle, i. 223a16–28. 9 Richard Sorabji, The Philosophy of the Commentators 200–600 A.D. (Ithaca, NY: Cornell University Press, 3 vols., 2005), ii. 203. 10 Ibid, 201–2 (parenthesis in the original). Richard Sorabji translates this passage directly from the Arabic translation of On Time by Ḥunayn ibn Isḥāq, published in ʿAbd al-Raḥmān Badawī, Shurūḥ ʿalā ArisṠū mafqūda fī al-Yunāniyya wa-rasāʾil ukhrā (Beirut: Dār al-Mashriq, 1971), 20–4. The Latin translation of the Arabic text by Grerad of Cremona in the twelfth century ce was translated by R. W. Sharples, ‘Alexander of Aphrodisias, On Time’, Phronesis, 27/1 (1982): 58–81. Acknowledging that this part is a commentary on Aristotle’s Physics 223a16–29, Sharples explains (p. 70) that ‘… both Aristotle and Alexander hold that time would not exist in the absence of soul. Alexander however, has two arguments for this conclusion where Aristotle has only one, explicitly. Both authors argue that time depends on soul on the ground that it is soul that numbers movement; but Alexander adds, apparently, as a separate argument, that in the absence of the soul there would not be any movement. It is his standard doctrine that the movement of the heavenly spheres is caused by the desire of their souls to emulate the changelessness of the Unmoved Mover. 11 Sorabji, The Philosophy of the Commentators, ii. 202. 12 Ibid. 13 Ibid. 14 Sharples, ‘Alexander of Aphrodisias, On Time’, 71. 15 Philoponus, Against Aristotle on the Eternity of the World (transl. Christian Wildberg; Ithaca, NY: Cornell University Press, 1987), 18–31. 16 Philoponus’s interest in refuting the eternity of heavenly bodies and motion may be regarded as a possible response to the Aristotelian paradox that Sharples pointed out. The reason is that the refutation of the argument for the eternity of heavenly bodies is an implicit refutation of the possibility of eternal time or time ‘in itself’. 17 Ibid, 41–57. 18 Ibid, 58–68. 19 Ibid, 69–76. 20 Ibid, 77–91. 21 Ibid, 93–121. 22 The argument against pre-existing matter similarly constitutes the kernel of Philoponus’s arguments in Contra Proclum against Proclus’s On the Eternity of the World. Proclus resorted to the Aristotelian reductio ad infinitum proof of pre-existing matter in order to prove the eternity of the world and refute the Christian arguments for the creation ex nihilo. See in particular Books 9 and 11 of Contra Proclum in John Philoponus, Against Proclus On the Eternity of the World 9–11 (transl. Michael Share; London: Duckworth Publishers, 2010). See also: Frans A. J. De Haas, John Philoponus’s New Definition of Prime Matter (Leiden: Brill, 1997), 1–45. 23 Philoponus, Against Aristotle on the Eternity of the World, 123–4. 24 Ibid, 124–5. 25 Generation is a form of motion, as Philoponus acknowledges. However, in this context he is primarily concerned with locomotion as the most fundamental form of motion as distinct from change in quality (alteration), change in quantity (growth/diminution), or change in substance (generation and corruption). He thus agrees with Aristotle on the primacy of motions. For example, see Aristotle’s Physics 260b29–261a13. 26 Philoponus, Against Aristotle on the Eternity of the World, 125–7. 27 Ibid, 127. 28 See chs. 2–6 in De Haas, John Philoponus’s New Definition of Prime Matter. 29 Ibid, 285–90. 30 See fragments 114–16 in Philoponus, Against Aristotle on the Eternity of the World, 128–31. 31 See fragments 121–2, in ibid, 133–7. See also al-Ghazālī’s reconstruction of this argument in the ‘First Discussion of the Pre-Eternity of the World’ in Incoherence of the Philosophers. Al-Ghazālī, in contrast to Philoponus, associates erring with the limitation of the human estimative faculty and imagination which mislead the intellect. See Abū Ḥāmid al-Ghazālī, Tahāfut al-falāsifa (ed. Sulaymān Dunyā; Cairo: Dār al-Maʿārif, 2000), 110–14. 32 See fragments 123–4 in Philoponus Against Aristotle on the Eternity of the World, 137–8. 33 See fragments 125–6 in ibid, 139–40. 34 See fragment 132 in ibid, 143–6. 35 See al-Kindī, Rasāʾil al-Kindī al-falsafiyya (ed. Muhammad ʿAbd al-Hādī Abū Rīda; Cairo, Dār al-Fikr al-ʿArabi, 2 vols., 1950–3), i. 98–106. 36 See al-Kindī, Rasāʾil al-Kindī al-falsafiyya, 193–4. 37 Ibid, 185–92. 38 Ibid, 199–207. 39 See al-Kindī, Rasāʾil al-Kindī al-falsafiyya, i. 117. 40 Herbert Davidson, ‘John Philoponus as a Source’, 372. 41 al-Fārābī, Kitāb al-Ḥurūf (ed. Muḥsin Mahdī; Beirut: Dār al-Mashriq, 2nd edn., 1990), 110–28. 42 Ikhwān al-Ṣafāʾ, Rasāʾil Ikhwān al-Ṣafāʾ wa-khillān al-wafāʾ (ed. Khayr al-Dīn al-Ziriklī; Cairo: al-Maṭbaʿa al-ʿArabiyya, 4 vols., 1928), i. 350. (My translation.) 43 al-Fārābī, Kitāb al-Fuṣūṣ (Hyderabad: Dāʾirat al-Maʿārif al-ʿUthmāniyyā, 1346/1927), 2. 44 See al-Kindī, Rasāʾil al-Kindī al-falsafiyya, i. 21. (My translation.) To sum up: the huwiyya of an object ‘x’ refers to its self-sameness or identity in perception; it is the principle of its individuation inasmuch as it is perceived by a perceiving soul. As such, huwiyya has an epistemic import. In contrast, māhiyya refers to the Aristotelian species-form, in abstraction from any individual characteristics; and mawjūd is a general term of an essentially metaphysical import referring to anything of which existence is predicated whether mental existence (wujūd dhihnī) or external existence (wujūd khārijī). 45 Ibid, 119. (My translation.) 46 al-Kindī, Rasāʾil al-Kindī al-falsafiyya, i. 167. 47 Adamson, Al-Kindī, 84–5. 48 Alfred Ivry, Al-Kindī’s Metaphysics: a translation of Yaʿqūb ibn Isḥāq al-Kindī’s treatise On first philosophy’ (Fī al-Falsafah al-ūlā) (Albany, NY: State University of New York Press, 1974), 26. 49 Jonathan Lear, Aristotle: the Desire to Understand (Chicago, IL: University of Chicago Press, 1999), 74. 50 R. P. Hardie and R. K. Gaye, The Complete Works of Aristotle, i. 251a24–7. 51 Jonathan Lear, Aristotle: the Desire to Understand, 82–3. 52 Alfred Ivry and Peter Adamson both translate al-wujūd as perception; the use of the word perception is rather misleading given that it is conventionally used in Western epistemology to refer to sense perception. As Abū Rīda explains in the introduction to his edition of the Rasāʾil, it should be translated as finding inasmuch as it derives from the Arabic verb to find (wajada). This suggestion is supported by al-Fārābī’s account of the genealogy of the use of the term in Arabic in general and in theoretical sciences in particular. Al-Fārābī argues that the general Arabic term refers to what a human being finds existing (kāʾin) in another being/human being. However, it can refer to being as such whenever it is used without qualification in theoretical sciences to translate the Greek (ɛ͗στιν)—to be. Hence al-Fārābī argues that mawjūd has to be translated as being in an objective sense independent of its reception by any human being. But since al-Kindī is qualifying wujūd in this context by ‘human’, it is more appropriate to translate it as finding rather than perception. See al-Fārābī, Kitāb al-Ḥurūf, 110–28. It is noteworthy that al-Kindī makes an innovation that is completely overlooked by Ivry and Adamson in their commentaries on this passage. Isḥāq ibn Ḥunayn in his translation of Physics 184a17–184b13 uses the word al-umūr meaning matters/things to refer to the things ‘which are more knowable and clear to us and proceed towards those which are clearer and more knowable by nature … ’. See Arisṭūṭālīs, al-Ṭabīʿa (ed. ʿAbd al-Raḥmān Badawī; Cairo: al-Dār al-Qawmiyya li-l-Ṭibāʿa wa-l-Nashr, 2 vols., 1964–5), i. 3. Abū Bishr Mattā uses the same Arabic word al-umūr to translate a similar passage in the Posterior Analytics I.2, 72a 1–5. See Arisṭūṭālīs, Manṭiq Arisṭū (ed. ʿAbd al-Raḥmān Badawī; Cairo: Dār al-Kutub al-Miṣriyya, 3 vols., 1948–52), iii. 314. In light of the results drawn from the last section, the importance of the use of wujūd as against umūr is that al-Kindī redefines the distinction between two modes of finding instead of two species of objects. However, in keeping with the terminology mostly employed in medieval philosophy, I will render wujūd as perception. 53 Ivry, Al-Kindi’s Metaphysics, 61. 54 See Aristotle, Physics, 184a17–184b13. 55 al-Kindī, Rasāʾil al-Kindī al-falsafiyya, i. 107. 56 Ibid, 107–8. 57 Ibid, 108–9. 58 Ibid, 167, 295. 59 Ivry, Al-Kindi’s Metaphysics, 135–6. 60 al-Kindī, Rasāʾil al-Kindī al-falsafiyya, i. 312–62. 61 Adamson, Al-Kindī, 127–8. 62 Ibid, 129. 63 al-Kindī, Rasāʾil al-Kindī al-falsafiyya, i. 372. (My translation.) 64 Adamson, Al-Kindī, 129. 65 Ibid. 66 Ibid, 133. 67 Ibid, 134. 68 Ibid. 69 See al-Kindī, Rasāʾil al-Kindī al-falsafiyya, i. 110–12. 70 Ibid, 301. (My translation.) 71 Peter Adamson, ‘Before Essence and Existence: al-Kindī’s Conception of Being’, Journal of the History of Philosophy, 40/3 (2002): 297–312. 72 Ibid, 302. 73 Ibid, 307. 74 Ibid, 308. 75 Ibid, 310. 76 Olga Lizzini, ‘Wuǧud/Mawǧud/Existence–Existent in Avicenna: A key ontological notion of Arabic philosophy’, Quaestio, 3 (2003): 111–38. 77 See Charles Kahn. Essays on Being. (New York: Oxford University Press, 2009), pp. 62–4. © The Author(s) (2018). Published by Oxford University Press on behalf of the Oxford Centre for Islamic Studies. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Islamic Studies Oxford University Press

Al-Kindī’s Argument for the Finitude of Time in His Critique of Aristotle’s Theory of the Eternity of the World in the Treatise on First Philosophy: The Role of the Perceiving soul and the Relation between Sensation and Intellection

Loading next page...
 
/lp/ou_press/al-kind-s-argument-for-the-finitude-of-time-in-his-critique-of-ehihkvPYd7
Publisher
Oxford University Press
Copyright
© The Author(s) (2018). Published by Oxford University Press on behalf of the Oxford Centre for Islamic Studies. All rights reserved. For permissions, please email: journals.permissions@oup.com
ISSN
0955-2340
eISSN
1471-6917
D.O.I.
10.1093/jis/ety022
Publisher site
See Article on Publisher Site

Abstract

Abstract The study presents a new interpretation of Abū Yaʿqūb al-Kindī’s (d. ca. 260/873) refutation, in the Treatise on First Philosophy, of Aristotle’s theory of the eternity of the world. Critiquing Herbert Davidson’s classical position that al-Kindī’s three refutations in the Treatise are reformulations of John Philoponus’s (d. 570) in the Contra Aristotelem, the study shows that while al-Kindī’s first and third proofs intersect with Philoponus’s the second one does not. The first part of the study examines the concept of perceptual being (huwiyya) and shows that al-Kindī’s second refutation of Aristotle’s theory of the eternity of the world, which Davidson considered superfluous, is based on a creative reading of Physics 223a16–28. There, Aristotle argues that time is a mind-dependent concept. Because time is dependent on the perceiving soul, al-Kindī distinguishes between time from the mathematical and metaphysical perspectives and time in connection with a body in motion, that is, time from a physical perspective. The second part of the paper, building on the first, offers a fresh perspective on the dichotomous relation between sensation and intellection in al-Kindī’s epistemology. It argues that this problematic relation, described recently by Peter Adamson as the ‘epistemic gap’ between sensation and intellection, could be explained in terms of an epistemic modality. Explaining that the distinction between time mathematically conceived and time physically conceived hinges on a more profound distinction between two epistemic modalities, the study reconstructs the relation between sensation and intellection in al-Kindī’s system. It concludes by indicating the way in which al-Kindī’s epistemic modality anticipates later developments in Islamic philosophy, specifically in the work of Avicenna. Prelude In this paper I offer a fresh interpretation of the second1 of the three counter-arguments Abū Yūsuf al-Kindī (d. ca. 260/873) puts forward in his Treatise on First Philosophy to refute Aristotle’s theory of the eternity of the world. I show that al-Kindī’s critique hinges on the relation his epistemology posits between perception, time and motion. I suggest that al-Kindī’s critique could be read as a creative reconstruction of Aristotle’s argument in Physics 223a16–282 for the connection between time and the soul, and of the reception of that argument by Alexander of Aphrodisias. On this basis, I question the long standing claim that al-Kindī simply reproduced Philoponus’s critiques of Aristotle’s and Proclus’s proofs of the world’s eternity. This claim has been the backbone of the mainstream interpretation of al-Kindī’s critique of Aristotelian metaphysics from the time it was first put forward by Herbert Davidson in 19693 to its recent espousal and reiteration by Peter Adamson.4 Building on the suggested interpretation of al-Kindī’s epistemological commitment to the relation between time, motion and perception, I go on to put forward a new position on al-Kindī’s division of the sciences, especially the distinction between physics and metaphysics. This position enables a new perspective, on the one hand, on the problematic relation between sensation and intellection in al-Kindī’s epistemology, and on the other, the closely connected problem of the relation between empirically particular claims in the domain of physics and universal claims in the domain of metaphysics. The first part of the paper establishes the background of al-Kindī’s critique, starting with the two senses of time presented in Aristotle’s Physics and his conception of the fourfold relationship among time, motion, finite bodies and the perceiving soul. It summarizes Philoponus’s key surviving criticisms in Book VI of the Contra Aristotelem,5 of Aristotle’s definition of motion and its implications for his argument for the eternity of the world. The paper reviews Aristotle’s argument in Physics 223a16–28 for the dependence of time as the measure of motion on the soul as the agent of that measuring. It then discusses Alexander of Aphrodisias’s endorsement of this Aristotelian position in his treatise On Time and its Arabic translation by Ḥunayn ibn Isḥāq (d. 260/873). Alexander’s treatise, as Sorabji and Sharples indicate,6 is one of the key texts in the Aristotelian commentary tradition that discusses in a sophisticated way Aristotle’s thesis on time and the soul. It was also translated around the time of al-Kindī. But I stress that the possible influence of Alexander’s treatise on al-Kindī’s epistemology is incidental to my concern with al-Kindī’s creative reconstruction of Aristotle’s arguments. Alexander’s work is adduced primarily to support the suggested interpretation of al-Kindī’s position on time and the soul. Subsequently, I turn to al-Kindī’s critique of Aristotle’s theory of the eternity of the world in section II of Treatise on First Philosophy. I question Davidson’s conclusion that al-Kindī’s proof was simply a reiteration of Philoponus’s, and Adamson’s commitment to that conclusion despite his indicating (more clearly than Davidson) the differences between the proofs of al-Kindī and Philoponus. I show that al-Kindī’s first and third refutations of Aristotle, which seem to coincide with Philoponus’s are consecutively mathematical and metaphysical, whereas the second is purely physical. I then suggest that al-Kindī’s second proof parallels and goes beyond Alexander’s defence of Aristotle’s argument (found in On Time) for the connection between time and the soul in Physics 223a16–28.7 Thereafter, I show how al-Kindī employs the inseparability of time and the soul to critique Aristotle’s reductio ad infinitum argument in Physics 251a9–28 on the basis of the impossibility of assuming a first or last change. I go on to show that, contrary to Philoponus’s mathematical argument for the finitude of time, al-Kindī’s second refutation of Aristotle’s theory of the eternity of the world is rooted in the view that time is finite inasmuch as it is the measure of motion by a finite perceiving soul. The second part of the paper reinforces the findings of the first part by responding to the challenges implied by the relation al-Kindī establishes between sensation and intellection. It starts with a consideration of the roots of this problematic relation in the opening methodological remarks of the second section of the Treatise on First Philosophy. It then turns to the ramifications of this problem in al-Kindī’s theory of knowledge more broadly, focusing mainly on Adamson’s thesis on the epistemic gap between sensation and intellection in that theory. I argue that al-Kindī’s position on time and the soul is a result of his commitment, expressed at the beginning of the Treatise, to keep the boundaries of the theoretical sciences (mathematics, physics and metaphysics) properly delineated. Al-Kindī holds that the key to this proper demarcation is to take note of the different ways each science corresponds to different modes of perceiving employed by distinct faculties of the soul. This argument supports the relation suggested in Part I between time and the soul in al-Kindī’s epistemology. It also helps connect his conception of time with the intimately related problem of the relation between intellection and sensation. Finally, in Part II of the paper, I try to situate the suggested interpretation of al-Kindī’s epistemology in the context of the broader concerns of early Islamic philosophy, focusing primarily on Avicenna. Part I: The Finitude of Time in al-Kindī’s Critique of Aristotle I.1 The two senses of time in Aristotle’s Physics IV and Alexander’s position In Physics IV.14, Aristotle writes: It is also worth considering how time can be related to the soul; and why time is thought to be in everything, both in earth and in sea and in heaven. It is because it is an attribute or state, of movement (since it is the number of movement) and all these things are movable (for they are all in place), and time and movement are together, both in respect of potentiality and in respect of actuality? Whether if soul did not exist time would exist or not, is a question that may fairly be asked; for if there cannot be someone to count there cannot be anything that can be counted either, so that evidently there cannot be number; for number is either what has been, or what can be, counted. But if nothing but soul, or in soul reason, is qualified to count, it is impossible for there to be time unless there is soul but only that of which time is an attribute, i.e. if movement can exist without soul. The before and after are attributes of the movement, and time is these qua countable.8 In this passage, Aristotle asks the question: if time is the measure of motion should we consider time a property of the substances in motion? But what if there is no one measuring? Would there still be measurement if there were no agent measuring? Would there be enumeration if there were no one enumerating? This simple question provoked much controversy among ancient commentators of Aristotle. Some, like Themistius, rejected the view that time does not have its own real existence (hypostasis). He described Alexander of Aphrodisias as ‘wrong and unfaithful to Aristotle’ for espousing this position.9 Others like Alexander supported this view; he writes in On Time: We say that without a soul to count motions there would be no time [lit: if the soul which counts motions came to nothing, time would come to nothing]. Neither if there were no time would the sphere be in motion. But if it were not in motion, all motions would come to nothing. For it [is the motion of the sphere] which accounts for [lit: is the reason for … ] all of them.10 In this passage, which belongs to the part of On Time where he expounds on Aristotle’s conception of time, Alexander explains that time, as Aristotle suggested, could be considered a quantitative category used by the perceiving soul rather than having an objective reality itself. However, Alexander is also aware that ‘Aristotle himself shows that the before and after in change, which are enumerable, can exist without soul, at least if there can be change without soul … ’11 So he argues that: ‘Perhaps the enumerable will not exist, but that which is contingently capable of being enumerable will exist, such as horses or men, but not the enumerable qua enumerable’.12 In defending Aristotle thus against the possible charge of idealism, Alexander holds that Aristotle’s claim that time depends on the soul does not mean that objects in motion do not exist if there is no soul capable of perceiving them in motion. Rather, Aristotle is saying only that time, that which measures change/motion, will not exist, but the objects in time will continue to exist. Hence, Alexander concludes that ‘ … if there were nothing to enumerate there would be no time. But nothing prevents the substrate of time, which is change, from existing’.13 So, while time as measurement would not exist if there were no agent carrying out the measuring, this does not negate the fact that the object still exists and that it is still in change/motion (kinēsis). Alexander’s defence of Aristotle's argument entails an interesting claim. He interprets Aristotle’s position as rendering time as measurement and not as an objective quality of the substance in change/motion. He implies that time defines the relation between the soul and the objects in motion. So, it is not that perceivable objects depend on the perceiving subject in Berkeley’s sense of esse est percipi. To the contrary, the object in motion will always exist. It is only time, qua the relation between the perceiving soul and the object in motion, that would cease to exist because there is no soul to measure motion and hence there is no relation to explore in the first place. Sharples raises another point related to the core argument of this study. He writes: There is, however, a problem. If time, as the numbered aspect of movement, is dependent on thought for its existence, how can a distinction still be drawn between time in itself as a continuum and time as divided in our thought? It does not seem likely that the explanation is to be sought in a contrast between our intellects, on the one hand, and a superhuman intellect which measures movement without introducing divisions into it, on the other. Rather, as Professor Sorabji has suggested to me, the explanation may be that the instant as it creates time is thought of as travelling, and while being so thought of cannot also be thought of as dividing. Or, putting what is essentially the same contrast in another way, the distinction may be between time as the continuous numerable movement of the heaven and divided time which is movement which is actually numbered.14 Sharples points out the paradox implied by Alexander’s and Aristotle’s position. If time is dependent on the soul, how could Alexander and Aristotle talk about time in itself as a continuous quantity that measures the movement of the heavens? The contrast here is between time as the totality of the actual instants dependent on the presence of a perceiving soul and time as a continuum in which everything changing exists. If time objectively exists as the measure of the continuous movement of the heavens, which have souls and are in motion, how can its existence depend on the finite measuring soul? Sharples suggests, citing Sarobji, that we can speak of continuous time theoretically but if it is not divided then it is not enumerable. So again, without finite souls perceiving similarly finite bodies, there is no enumerable, divided time, the only kind of time that is our concern. Apart from the contrast Sharples draws between Alexander’s position and that of Galen’s to whom he was reportedly responding, the philosophically significant point here is how to reconcile these two positions on time? Further, would such reconciliation still allow for the possibility of talking about time ‘in itself’, the way that Alexander talked about it? I.2 Philoponus’s critique of Aristotle’s theory of the eternity of the world In order to appreciate the subtle relation between al-Kindī’s critique of the eternity of the world and Philoponus’s critique of the same argument, an overview of Philoponous’s arguments is necessary. The surviving fragments from Books I–VI and VIII of Philoponus’s Contra Aristotelem have been pieced together from Simplicius’s commentaries on Aristotle’s Physics and De Caelo in addition to Arabic sources, mostly the works of al-Fārābī (d. 338/950) and al-Sijistānī (d. ca. 390/1000).15 Philoponus devoted Books I–V to his refutation of Aristotle’s arguments for the eternity of the world in the De Caelo and the Meteorology.16 Book I criticizes Aristotle’s theory of the fifth element in De Caelo I.2 claiming that the motion of fire is also circular. While he agrees with Aristotle that motion is determined by the ‘nature’ of the elements, he disagrees that circular motion should be regarded as eternal. This step enables him to question the eternity of the celestial bodies or the sensible non-perishable substances of Aristotle.17 Book II continues with the critique of De Caelo 1.3 269b18–270a12 concerning the weight of the heavens, arguing that the latter do have weight and are consequently natural substances.18 Book III turns to Aristotle’s Meteorology arguing, in line with Plato’s position in the Timaeus, that the heavens are mostly made of fire and hence are not eternal or made from a fifth element.19 In Book IV Philoponus argues that Aristotle’s argument for the eternity of the heavens in De Caelo I.3 270a12–22 is dialectical and not demonstrative.20 Book V mainly argues against the Aristotelian position in De Caelo I.4 that circular motion has no contraries and hence is eternal.21 It is in Book VI that Philoponus turns to the relation between motion, time and sensible substances, directly identifying problems with Aristotle’s conception of that relation. The extant arguments from Book VI can generally be divided into two groups. The first group critique Aristotle’s view that motion is the actualization (energeia) of pre-existing matter. Aristotle argued reductio ad infinitum that matter must be eternal because if matter were generated, then a potentially pre-existing matter must be posited, leading to an infinite regress. This group of arguments constitutes the core of Philoponus’s critique of Aristotle;22 owing to their focus and scope, I will call them ‘physical arguments’. The second group of arguments do not have a central theme; however, they are mostly logical/metaphysical and mathematical. I will thus refer to this group of arguments as ‘speculative and mathematical arguments’. In the fragment from Book VI preserved in Simplicius’s Physics 1129.28–1131.7, Philoponus writes: When Aristotle defined motion as the actuality of the moveable qua moveable, he covered by this definition … all motion in general, yet he assumes that some motions are eternal whereas others possess a beginning and an end. Now, what are his grounds for assuming as a consequence of the definition of motion that things that are going to be moved necessarily pre-exist in time the non-eternal motion, which possesses a beginning of existence, as things that possess only the capacity (dunamis) of motion without the actualization (energeia)?. . Every definition … is predicated of all things defined with equal validity … it follows that from the definition that in the case of non-eternal <motion> the moved object must pre-exist the motion in time, the same will follow in the case of the <eternal motion> too. If this is true the definition is necessarily either (i) not valid … in the case of eternal motion … which is paradoxical. Or (ii) if it were true also in the case of <eternal motion> that the moved object pre-exists the motion in time—so that too the substance of the heavens pre-exists the circular motion in time—<and if> none of the things that possess something pre-existing in time before them is eternal, then eternal motion in this sense will not be eternal … Or (iii) in case they want motion to be eternal, <then> it is not true that it necessarily follows from the definition of motion that the movable pre-exists the motion in time.23 The core of Philoponus’s argument could be paraphrased as follows. Aristotle defines motion as the actualization of the potentially moveable object, implying that the potentially moveable object pre-exists motion. If this definition is correct, then eternal motion would not be possible. For if this assumption were granted, there would be something that precedes the eternal, which is a contradiction a priori. To solve this problem, Philoponus argues: … that when Aristotle gave the definition of motion he did not have the same idea (ennoia) of the potential (to dunamei), <i.e.> as something that can be separated from the actualisation (energeia), that is to say, the motion; but <his idea was> that motion exists as long as the potential is present … and that the potential exists as long as there is motion. For when the motion stops whenever the moved object has reached its completion (telos), then the potential stops as well … Aristotle understood ‘capacity’ (dunamis) as being together with motion and not as <anything> separate from the actualization … 24 In order to overcome the paradoxical implications of Aristotle’s definition of motion, Philoponus re-interprets the relation between potentiality and motion. Both are inseparable. There is nothing that has the potentiality to move and then is actualized through motion. Rather, that which is moveable has the potential to be moved as long as there is motion. The end of motion implies the fulfillment of the potential to move. Philoponus restricts the potentiality and actuality of motion to the interval in which motion/change takes place. This restriction on the concept of motion enables Philoponus to resolve the paradox of eternal motion implied by Aristotle’s definition even though he will refute the possibility of eternal motion later. However, he is interested in pointing out the contradictions entailed by Aristotle’s definition of motion. Philoponus applies the same theory to non-eternal motion. He argues that wood does not possess the potentiality of moving upwards when it is burnt. When wood is burned in fire; this phenomenon amounts to generation and not motion.25 The upward motion of fire is not the actualization of the wood; wood is heavy and naturally moves downwards in contrast with fire which naturally moves upwards.26 It would be absurd to claim that the upward motion of fire resulting from the burning of the wood is the actualization of the latter. Hence, the potentially moveable does not pre-exist the actually moving. Rather, potentiality/capacity (dunamis) is simultaneous with motion/actuality. … motion is the actuality of the immediate capacity. So if fire … and not wood is the potentially movable, and if it is no sooner generated than on the move towards the upper <region>, unless some force prevents it—and similarly the water in the clouds—then it is not true <to say> that in the case of non-eternal motions the potentially movable pre-exists the active motion in time.27 Philoponus manages in this way to undermine Aristotle’s argument for the eternity of motion. For, if no potential thing temporally pre-exists that which is moved/actualized, then the conclusion Aristotle draws that motion and its measure, viz. time, is eternal does not necessarily follow. But an even more important conclusion follows from examining the first set of Philoponus’s arguments against Aristotle’s definition of motion. Philoponus establishes a synchronic inseparability between time, motion and sensible substances. This synchronic relation hinges on a careful consideration of the actuality of motion of sensible substances in experience. It avoids as much as possible any theoretical assumptions. Philoponus bases his physical arguments on the actual physical experience of the different forms of motion that sensible substances exhibit in concreto. This point is in harmony with Frans De Haas’s study of Philoponus’s conception of prime matter as three-dimensional body. As indicated above, Philoponus accepted the principle of concreatio of matter and form, based on the inseparability of form and matter and the impossibility of conceiving incorporeal material substances (that is substances that are material but are not corporeal and hence not three-dimensional).28 Hence, Philoponus argued that if there is any prime matter it is some indeterminate three-dimensional body. This position, which avoids as much as possible theoretical assumptions about the physical world and focuses on the examination of sensible substances as they are actually perceived in finite experience, allows Philoponus to critique Aristotle’s theory of the fifth element. According to Philoponus celestial bodies share obvious characteristics with sensible sublunar substances. Hence, no reasonable ground warrants the assumption that they are composed of a non-perishable fifth element. As De Haas rightly explains, Philoponus wants to confine physics to objects in motion and avoid invoking theoretical speculation about the principles of the science like those about prime matter to explain motion. In so doing, he is more loyal than Aristotle to Aristotle’s own principle found in the Posterior Analytics of not questioning the principles of a science within the science in question.29 Turning to Philoponus’s ‘speculative and mathematical’ set of arguments against Aristotle’s theory of the eternity of the world, I will summarize three of them. The first centres on the critique of the statement, ‘nothing comes to be from nothing’. According to Philoponus this theory logically applies to nature but not to God; otherwise God would not be above nature as required by definition.30 Subsequently, Philoponus refutes three major speculative arguments Aristotle employs to prove the eternity of the world. The first of these is that there must always be time because there is always before and after. In response, Philoponus argues that our tendency as humans to associate what is before with the past and what is after with the future is due to the limitations of the human intellect.31 Angels and God both have the notion of before and after without being in time. Aristotle’s second argument is that almost all physicists are of the same opinion. Philoponus replies that truth is not determined democratically but rather by intellectual enquiry, giving his preference to Plato.32 The third of Aristotle’s arguments is based on the concept of the ‘now’. Since time cannot be conceived without a concept of now and since now is a mean between the past and the future there will always be a time on both sides of now; hence, time is eternal. Philoponus responds by showing that Aristotle commits petitio principii by assuming time in his conclusion. The assumption that ‘now’ is a mean of time is the same as the claim that time is eternal.33 Finally, Philoponus gives three lengthy mathematical arguments against the eternity of the world. The most relevant of these is that if time was eternal then it would be impossible to traverse from an infinite past to a specific point in time. Hence, nothing could be generated from something that is infinite.34 The key characteristic of Philoponus’s second set of arguments against Aristotle’s theory of the eternity of the world is its metaphysical/mathematical nature. On the one hand, Philoponus puts forward a number of speculative arguments like that of the power of God to create ex nihilo by definition. On the other hand, he puts forward a set of mathematical arguments demonstrating that it is mathematically impossible to have anything generated if motion and hence time are infinite. Turning to al-Kindī, I will show that he goes considerably further than Philoponus’s empirical commitment in studying physical phenomenon by examining the epistemic condition of physical experience. I will demonstrate that he expands the role of the perceiving soul not only as the condition of the existence of time, but also as the epistemic basis for determining the different domains of theoretical knowledge and relations among them. This development, as I will argue in the last section, can shed fresh light on the problematic implications of the existential relation between the soul and time that Aristotle and Alexander posited. I.3 Between Philoponus and Alexander of Aphrodisias: al-Kindī’s second proof of the finitude of the world In the second section of On First Philosophy, al-Kindī introduces a number of key methodological distinctions on the boundaries and methods of metaphysics, physics and mathematics. I will return to these methodological remarks in the next section. It suffices here to mention that al-Kindī insists that physical phenomena should not be approached from a mathematical perspective or a logical/metaphysical perspective.35 Subsequently, he foregrounds his proofs of the finitude of the world with a series of definitions and axioms concerning the eternal and the possibility of infinite bodies. Al-Kindī starts, in a manner reminiscent of Philoponus, with a mathematical demonstration that any sensible substance (jirm) must be finite. He follows the same logic in other treatises including ‘A treatise on the essence of what cannot be infinite and what could be called infinite’,36 ‘A treatise to Aḥmad ibn Muḥammad al-Khurāsānī on the finitude of the world body’37 and ‘A treatise to ʿAlī ibn al-Jahm on the unity of God and the finitude of the world body’.38 He then provides a series of logical proofs to demonstrate the impossibility of arguing for an infinite body. For instance, he argues that if there were an infinite body and we imagine separating part of it, the remainder will be either finite or infinite. If the remainder is finite, we must conclude by reductio ad absurdum that no infinite bodies could possibly exist because the sum of two finite quantities must be finite. If the remainder is infinite and if we try to add to it the part we separated, the sum would be the same; the infinite would remain infinite after separation. Since this result does not follow the rules of quantity as applied to possible bodies, then no quantity could be posited as infinite. But this concerns bodies qua quantities, that is from a mathematical perspective. Judging whether or not these bodies can be in finite or infinite motion and hence whether time, qua the measure of motion, is eternal or not, is a question for physics that studies being qua motion, as Aristotle instructs in Book I of the Physics; it is not a question for metaphysics or mathematics. ‘Motion’ is used in the technical Aristotelian sense throughout this paper, the sense it has in Book I of Physics. In discussing Philoponus, I explained the four kinds of motion or change, and the primacy of locomotion. Hence, al-Kindī embarks on a physical inquiry. Following Aristotle in Book IV of the Physics, al-Kindī, approaches the problem of the infinity of time based on a threefold relation between body, time and motion. He starts by asserting that there is no body without time because time is the measure (number) of motion. Motion in turn is not possible except through a body. Al-Kindī writes: If there is a body there is motion otherwise there would not be motion. For motion is some kind of change. For instance the change of the parts of a body or its centre or all its parts is locomotion … there is also growth and diminution … change of qualities is alteration and the change in substance is generation and corruption … change is thus attributed to that which is temporal (dhī al-zamān). Hence if there is motion then there is body … 39 Nonetheless, al-Kindī does not rule out the possibility that time is infinite on a potential level. He does not deny that, mathematically, time is a continuous quantity and hence could be thought of as infinite. However, this is not al-Kindī’s major concern in the context of establishing the relation between time and sensible substances from the perspective of physics; quantities are the object of mathematical inquiry. In this regard, he is mainly concerned with the nature of sensible substances and the legitimacy of the knowledge claims that one can make about them. From this perspective, time, motion and bodies are inextricably connected, each implies the other as far as their possibility in experience is concerned. In order for any sensible substance to be such, it has to be in motion and since time cannot precede motion because the latter is its measure, then the three categories, body, motion and time, all imply each other. Both Davidson and Adamson consider al-Kindī’s position in this passage a reiteration of Philoponus’s physical arguments. However, a careful examination reveals otherwise. As I have shown above, Philoponus’s argument for the inseparability of time, motion and sensible substances is based on his conception of capacity/potentiality (dunamis) as inhering in the moveable object as long as it is moved; the end of motion is equivalent to the actualization of the potentiality. It is true that al-Kindī may have been aware of this Philoponian position from the Contra Aristotelem and the Contra Proclum. But he introduces what Davidson considered a superfluous step ‘ … inferring from the finiteness of time that the ‘body’ of the universe is finite ‘in its being’, and only then, from its finiteness ‘in being’, concluding that the ‘body’ of the universe cannot have existed forever’.40 The term al-Kindī uses for the ‘being’ of the sensible substance is huwiyya. But what does al-Kindī mean by the huwiyya of the sensible substance in this context? The pronoun huwa and the verbal noun (maṣdar), huwiyya, were among the earliest Arabic terms used to denote, respectively, a being and the being of this being. In Kitāb al-Ḥurūf al-Fārābī explains that the pronoun huwa was selected to render the word being/substance because it typically serves as the subject of the sentence, in line with Aristotle’s characterization of substance as the primary subject of predication in the Categories. He explains also that it had been at one point preferred to what became the standard term for being, mawjūd. The reason was the fear of Arab translators of the many implications of mawjūd, a derivative of wajada (to find, etc.). Al-Fārābī, adds that huwa and huwiyya also had a demonstrative/existential sense inasmuch as both indicate to the object perceived.41 In the Epistles of the Brethren of Purity, the term huwiyya is used for perceptual being, as against māhiyya which was used to indicate essence: ‘The perceptual beings (huwiyyāt) of things are acquired by the souls through the senses whereas the essences (māhiyyāt) are acquired by the intellect.’42 Similarly, al-Fārābī in Kitāb al-Fuṣūṣ argues that: ‘the huwiyya of the human being is animal nature and corporeality … ’43 where huwiyya refers to accidents versus the māhiyya—in this case rationality—which refers to the distinctive essence of the human being. Davidson, after textual analysis of of al-Kindī’s second proof, considered it superfluous, due to its complex composition. I will argue that al-Kindī uses the term huwiyya to indicate perceptual being. I mean by perceptual being the being of a being as long as it is perceived by a perceiving soul. Al-Kindī systematically insists that the being of every sensible substance is found within a specific temporal frame of reference (mudda) in which it is what it is. Abū Rīda suggested this meaning (huwiyya) more than half a century ago. In the introduction to his edition of al-Kindī’s treatises he writes: ‘The substantive al-huwiyya is derived from the pronoun al-huwa; the former means the individuated (mutaʿayyin) existence that is grasped by the senses in contrast with the truth or essence of the sensible substance which are grasped by reason’.44Huwiyya and the verb al-Kindī associated with it (yatahawwā) mean to exist to the senses. It denotes the existence (kawn) of the sensible substance in perception; but in order for perception to occur there must be a perceiving soul. Hence, huwiyya denotes the existence of sensible substances in perception that is relative to a perceiving soul. To elucidate this point, I will consider the same passage Abū Rīda uses to articulate his interpretation of huwiyya. In Part II of the Treatise, al-Kindī peculiarly argues: It has been established above that time does not precede motion. Time by necessity does not precede the body (al-jirm); for there is no time without motion [and only bodies are capable of motion]. Hence, no body exists without motion and no motion exists without body and no body exists except within a temporal frame (mudda); this temporal frame is period in which a body is a huwiyya, i.e., the temporal frame is the period in which the body is its huwiyya.45 The gist of al-Kindī’s argument may be reconstructed as follows. Only bodies can be in motion and all bodies are finite. But time is the measure of motion. Hence, time cannot be infinite. In this way, for al-Kindī time, motion and body are inextricably related to each other in the construction of experience. It is untenable, from the perspective of physics or study of being qua motion, to claim that time is infinite since time is the measure of motion and only bodies can be in motion and all bodies are finite. But Aristotle already makes this argument. How does al-Kindī differ from him? The answer to this question hinges on the connection al-Kindī draws between time, motion and bodies through the concept of huwiyya. As explained above, huwiyya refers to the being of a sensible substance in perception (idrāk). Al-Kindī, like early Muʿtazilī thinkers (whose work he knew well), and also in line with their later Ashʿarī rivals, does not reduce the reality (ḥaqīqa) of a thing to its definition (ḥadd) or conception (mafhūm); rather the reality of a thing consists in the definition and conception of it as embodied in its particular, physically perceptible existence. Time measures the period (mudda) in which a thing is what it is (huwa mā huwa)—that is the period in which a material substance is its reality. This reality is essentially the perceptible being of the material substance in concreto. Thus huwiyya is essentially temporal since perception is always in time. This is different from the general Aristotelian theoretical claim that sensible substances and time are connected through motion. Al-Kindī makes a more exacting empirical claim. He is not merely following Aristotle and Philoponus in claiming that there is motion if and only if there are sensible substances. He establishes an immediate connection between bodies and time. He writes, ‘there is no body without a temporal frame’ (lā jirm bi-lā mudda). This statement cannot be interpreted to mean that bodies would not exist if there were no time intervals in which they existed. Rather, it means that bodies are in time because they always are what they really are for a specific interval. This interval has to be in reference to something. I would like to claim that this ‘something’ in reference to which the interval is determined is the perceiving soul. In order to further substantiate this claim, let us consider how al-Kindī’s claim may be better understood in terms of our account of Alexander’s defence of Aristotle’s claims about the dependency of time on the perceiving soul. The core of al-Kindī’s argument concerning the relation between time and sensible substances consists in his insistence on limiting this relation, as Philoponus does, to the boundaries of experience. A sensible substance is always determined as such within a temporal frame, the duration in which it is perceived by a perceiving soul. Perception furnishes access to the physical reality of material/corporeal substances/things. In turn, the reality of anything is the particular, material instantiation of the definition of this thing in its specific individual being; the reality of any human is the definition of the human as instantiated in this particular individual human with all their particular attributes. The perceptual determination of the reality of a sensible substance within a specific temporal frame is denoted by huwiyya. Therefore, the huwiyya of a sensible substance is correlated to the soul being sensible of it; this correlation is determined in terms of time. If this is so, time is constitutive of huwiyya; time cannot merely be an objective measure of motion for the physical reality of the thing is always accessed in time. Hence, al-Kindī gives a rather peculiar definition of time in his Risāla fī ḥudūd al-ashyāʾ wa-rusūmahā ‘Treatise on the Essential/Complete and Inessential/Incomplete Definitions of Things’ (rasm is definition using non-essential differentia, whereas ḥadd is real definition using the essential differentia): ‘Time: is a temporal frame (mudda) measured by motion … ’46 Time does not measure motion; motion measures time. I take this definition to refer to the direct relation al-Kindī wants to draw between time and the perception of the sensible substance. Sensible substances are always perceived in time; time is constitutive of perception. Time cannot be conceived independently of the sense-experience of a specific material substance. Since every sensible substance actually in perception is always in motion and in time in reference to a perceiving soul, there is absolutely no ground for claiming that the world is eternal. Such a claim would be merely speculative and by no means satisfies the conditions of physical investigation. Time defines the relation between the perceiving soul and the sensible substance; such relation is always finite because perception is finite. This argument reinforces, from the perspective of physics, al-Kindī’s previous mathematical proof of the impossibility of an actually existing infinite sensible substance. However, its real value consists in articulating the relation between time and the perception of sensible substances. Against this background it can be argued that the step Davidson considered superfluous is actually rather meaningful, especially as far as it sets al-Kindī’s second proof apart from Philoponus’s physical proofs. Time is finite because it is one of the categories through which a perceiving soul perceives a sensible substance. Based on this and ‘ … on this alone’, as Davidson contends, the sensible substances and their totality, that is the world, must be finite. Sensible substances are always perceived in time and hence in motion and since the former is finite the latter must also be so. But how is this position different from Philoponus’s physical argument? As I indicated in the previous section, Philoponus’s physical argument mainly depends on his adherence to the Aristotelian principle of physics as the study of being qua motion. Avoiding theoretical/speculative assumptions like the positing of pre-existent matter, Philoponus reconstructs Aristotelian concepts of motion and the relation between form and matter in light of actual, observable, finite experience. While, al-Kindī makes a parallel move, his focus on the relation between time, motion and the soul, as against Philoponus’s focus on reconstructing the concept of motion, relocates the critique of Aristotle’s theory of the eternity of the world on different ground. It is true that al-Kindī’s first and third arguments against the eternity of the world parallel Philoponus’s mathematical critiques of the eternity of the world. For instance, in the third argument al-Kindī follows Philoponus’s mathematical argument closely as has been shown by Herbert Davidson, Alfred Ivry and, more recently, Peter Adamson. Adamson calls this argument ‘the counting argument’ and summarizes it as follows. According to Philoponus ‘ … were the world to be eternal ex parte ante then we would be committed to an actual infinity, for example the number of years that have already elapsed. Worse still, we would have an actual infinity that is still increasing. For example, the number of years that had elapsed when Socrates was alive has increased since then, so we would have one actual infinity that is bigger than another, which is (according to both Philoponus and Aristotle) impossible’.47 However, al-Kindī’s argument against the infinity of time that I have explained above, and which Adamson agrees is the longest and most important one in his critique of Aristotle’s argument for the eternity of the world, does not owe much to Philoponus. But if al-Kindī merely echoes Aristotle’s position that time’s existence depends on the soul, a position already expounded by Alexander, what, if any, new contribution did he make to the medieval debate on the eternity of the world? More importantly, does al-Kindī’s position (assuming that my interpretation is correct) help illuminate any further aspects of his philosophy and its contributions to the history of philosophy at large? To respond to the first question, I will turn to the second part of this section. Peter Adamson and Alfred Ivry pointed to a rather paradoxical position that al-Kindī takes in complete opposition to Philoponus. Despite the fact that al-Kindī argued that the world, space and time are finite, he accepted Aristotle’s argument regarding the mathematical/potential possibility of the eternity of the world. He also accepted Aristotle’s argument in the De Caelo for the eternity of the supra-lunar sphere. Ivry writes: ‘[al-Kindī] … accepts the Aristotelian description of the fifth element as simple, ceaselessly moving substance; and agrees with Aristotle’s description of the supra-lunar spheres as not having generation and corruption, being perfectly circular and concentric … Al-Kindī seems to be saying that the world, though not eternal, is in other respects as Aristotle said it was; except it need not be so and would not be so, were it not for God’s will’.48 This paradox conjures up the one Sharples claimed is implied by Alexander’s position in On Time: how can the same thing, time in this case, be deemed as an infinite continuous quantity and at the same time an actual, finite instant? Al-Kindī’s position on the relation between time and the soul affords him ample flexibility in dealing with the two senses of time Aristotle identifies. As Jonathan Lear argued, Aristotle’s account of time operates on two levels: ‘First he gives a theoretical understanding of the nature of time. Second, he wishes to account for our experience of time. These two levels are related.’49 In Aristotle’s Physics VIII.1 251a24–7, Aristotle writes: For if we are to say that, while there are on the one hand things that are moveable, and on the other hand things that are motive, there is a time when there is a first mover and a first moved, and another time when there is no such thing but only something that is at rest, then this thing must previously have been in process of change; for there must have been some cause of its rest, rest being the privation of motion. Therefore before this first change there will be a previous change.50 This passage clearly indicates the contrast between al-Kindī and Aristotle. Aristotle bases his argument for the eternity of world and the infinity of time on the assertion that there are no bodies without motion and that rest is the privation of motion. This is a logical argument based on a theoretical assumption about the concept of motion. Lear comments on this passage as follows: Aristotle is arguing that given any purported first change, there must have been a change, which existed before it. (In a similar vein he argues that given any purported last change, there must be a change which occurs after it.) Thus we can understand his claim that there has always been change as being more than an analytic truth, if we interpret him as claiming that it is absurd for there to have been a first change. Similarly, Aristotle’s claim that the world is eternal should not be interpreted in terms of an infinitely expanded length of time, but only as a claim that no moment could be first or last moment of the world’s existence. Aristotle’s argument establishes no more than the potential infinity of time: time is such that for any moment in time it is possible to find an earlier and a later moment … The fact that infinity of time is only potential is intimately bound up with the role that mind plays. An actual infinite extension of time would, in Aristotle’s eyes, do nothing for us: for we are beings who cannot possibly comprehend an actual infinite extension. But it is the very essence of time to be comprehensible to us. Since the very reality of time is manifested in the soul’s measurements, the infinity of time can be grounded in nothing more than the fact that, given any change, it will always be possible to measure an earlier and a later change … Because our experience of time partially constitutes its reality, Aristotle can infer from our experience of time to the very nature of its existence.51 Lear’s argument in defence of Aristotle’s argument for the infinity of time could be summarized as follows. Aristotle is arguing for the potential infinity of time, which is logically plausible given his assumption that there is no motion without bodies and that bodies are by nature in motion. Rest is a privation of motion; so if a first motion and a first mover are posited infinite regress necessarily follows. Hence, time has to be infinite and the world has to be eternal. However, Aristotle also realizes that we are incapable of comprehending actual infinity; as rational beings we can only posit it potentially. Furthermore, Aristotle recognizes the dependence of time’s existence on the rational soul. Since this is the case we can infer that time is infinite since we are obliged to think of time’s potential infinity. Therefore, it is plausible to infer that time is infinite and that the world is eternal. Al-Kindī’s position has an epistemic advantage over that of Aristotle. If time depends on the soul inasmuch as it perceives a sensible substance, then all assertive claims about time have to be restricted to the finitude of such experience. But in thinking about time on a mathematical level as a continuous quantity abstracted from the world, another set of assertions may be warranted. The problem with Aristotle’s argument, especially in light of Lear’s exposition of it, is that it expands logical and mathematical arguments to the domain of actual, physical experience. The supra-lunar sphere is not sensibly experienced; therefore some speculation can be allowed about its nature. It is in this domain that God’s intervention is allowed and conceived of on a strictly logical and rational basis inasmuch as God is purely transcendent to the world. Al-Kindī radically distinguishes physical arguments from logical or metaphysical ones on the one hand and from mathematical ones on the other hand. For al-Kindī, time, as a continuous quantity, is understood from a mathematical perspective as infinite. This theoretical claim which establishes the ground of the potential infinity of time, as indicated by Lear, must not be conflated with or brought to bear on any judgment concerning the physical nature of the world. The principle of physics is motion, as Aristotle himself systematically asserts in Book I of the Physics. Al-Kindī strictly adheres to this principle. If empirical exploration of sensible substances in motion has revealed to us the essential connection between time, motion and bodies and time has been shown to be connected with the perceiving soul, then the eternity of the world cannot be warranted. It is merely a theoretical claim that has to be distinguished from any assertion about the nature of the physical world. This epistemic distinction affords al-Kindī the possibility of dealing with time differently in accordance with the different principles of each domain of investigation. The difference in principles, as we learn from the Posterior Analytics, respectively determines the approach to beings within each domain. This leads to the second question as to whether this position on time and the soul reflects a deeper commitment in al-Kindī’s philosophy and whether this position makes an interesting contribution to Islamic philosophy at large. Part II: Separating Physics and Metaphysics: The Opening Remarks of Section II of On First Philosophy II.1 Problematizing the relation between sensation and intellection in al-Kindī’s epistemology In the previous section I argued that al-Kindī’s position on time is inextricably connected with his strict division of the theoretical sciences: mathematics, physics and metaphysics. Al-Kindī’s physical argument, as distinct from his other mathematical and metaphysical positions, against the eternity of the world hinges on the epistemological connection between time, motion, corporeality and the soul in concreto. In this section, I further explore the roots of this fundamental division of sciences in the epistemic distinction al-Kindī draws between sensation and intellection. I will show how this epistemic distinction or ‘gap’, as Peter Adamson labels it, indicates the central role al-Kindī assigns to the different and essentially distinct roles of the faculties of the soul in his epistemology. Articulating the centrality of this role has a twofold objective. On the one hand, it reinforces the interpretation argued above of the relation between time and the perceiving soul; on the other, it briefly situates al-Kindī’s epistemological position on the division of the sciences and its roots in the distinct functions of the different faculties of the soul in the broader context of the interests of early Islamic philosophy. I will start with a brief consideration of al-Kindī’s opening remarks in the second chapter of the Treatise and how they fit in a larger context of the problematic relation between sense and intellection and hence physics and metaphysics. Al-Kindī peculiarly opens the second section of the Treatise which he subtitles ‘The first part of First Philosophy’ or the first part of metaphysics with a rather creative reformulation of Aristotle’s Physics 184a17–184b13. Al-Kindī writes: [T]here are two kinds of human perceptions,52 one of which is nearer to us and further from nature. This is the perception of the senses, which belong to us from the beginning of our development, and belong to the genus common to us and to many others, i.e. life, which is common to all animals. Our perceiving with the senses, through the contact of the sense with its sensible object, takes neither time nor effort, and it is unstable, due to the motion and fluctuation of that which we contact, its change in every case being through one of the kinds of motion. Its quantity is differentiated by ‘more’ or ‘less’, ‘equal’ and ‘unequal’, while its quality is contrasted by ‘similar’ and ‘dissimilar’, ‘stronger’ and ‘weaker’. Thus it always occurs in continuous motion and uninterrupted change. It (sc. sensory perception) is that the forms of which are established in the imagination, which conveys them to the memory; and it (sc. the sensible object) is represented and portrayed in the soul of the living being. Though it has no stability in nature, being far from nature and therefore hidden, it is very near to the perceiver, in that his perception is due to the sense, with the contact of the sense with it (sc. the sensible). All sensibles, moreover, are always material, and the sensible is always body and in a body. The other finding/perception is nearer to nature and further from us, being the finding/perception of the intellect.53 In contrast to the same passage in the Physics where Aristotle advises that in any enquiry we should start from ‘the things, which are more knowable and clear to us and proceed towards those which are clearer and more knowable by nature … ’54 al-Kindī advises that we should start from the mode of perception that is closer to us. This mode of perception is immediate sensation that does not require time or effort to discover inasmuch as it is always readily present to us, in contrast to rational perception by the intellect which is more remote from us but closer to nature.55 In other words, al-Kindī stresses that the distinction is between two modes of epistemic comportment associated with two different faculties of the soul rather than laying the emphasis on the distinction between two different objects of knowledge. Having preliminarily distinguished between these modes of human perception, al-Kindī gives a further distinction between them. Sense perception is always representational (mutamaththil). All representations (muthūl) are of sensible qualities (maḥsūsāt); hence, they depend on sense perception.56 By contrast, objects of the intellectual perception are not representational but recognized of logical necessity independently of any representation.57 Representation happens through the internal senses (al-ḥawāss al-bāṭina), specifically through imagination. Imagination in this context is completely reduced to a function related to the re-presentation of substantial material forms that constitute the particularity of every individual substance. Al-Kindī avoids any discussion of the intellectual dimension of imagination in mediating these forms to reason in contrast with his definition of φαντασια in On Definitions and On the Essence of Sleep and Vision.58 Alfred Ivry indicated this point in his commentary on this passage from the Treatise on First Philosophy: It is, in fact, the ‘sensational’ aspect of imagination … , which animals share with men, to which al-Kindī is particularly referring in this sentence, apparently ignoring its more ‘rational’ or ‘deliberative’ character … and it is the latter function which in man prepares his perceptions for comprehension by the intellect. That al-Kindī was familiar with this rationalizing role of the imagination is evident from his definition of tawahhum (ordinarily rendered as ‘estimation’ but used by al-Kindī as the equivalent of ‘imagination’ in the broad sense) in On Definitions … ‘it is … a psychic faculty which apprehends sensory forms in the absence of their matter. It is also said that … imagination is the presenting of the forms of sensible things in the absence of their matter’ … This point is repeated in his treatise On the Essence of Sleep and Vision … Thus it is probably not insignificant that al-Kindī here chooses to refer to a view of sense perception which ostensibly has nothing to do with the intellect; and that he mentions two of the ‘internal senses’ but omits the third and, for man, the most important, viz., the rational faculty, in which perceptions normally culminate. This omission may be due to his desire to draw a sharp contrast between sensory and intellectual perceptions; and to construct a philosophy along lines that are ‘purely’ logical and demonstrative.59 Ivry explains al-Kindī’s interest in focusing on the sensible dimension of the imagination in terms of his ‘ … desire to draw a sharp contrast between sensory and intellectual perceptions’. But he does not explain why al-Kindī is interested in drawing this sharp distinction. Adamson has recently given a rather complex analysis of this contrast. In Al-Kindī, Peter Adamson discusses the contrast between sense perception and rational consciousness at some length. Against the background of his discussion of al-Kindī’s psychology, especially in ‘On the Intellect’,60 Adamson defines two methods of interpreting the complex relation between sense perception and intellection which he respectively describes as the weak thesis and the strong thesis. Adamson describes these two methods as follows: … it will be useful to distinguish between two ways of making intellection, and not sensation, the faculty through which we have knowledge. First, one might hold the relatively weak thesis that sensation and intellection have different objects and that knowledge in the strict sense is only of the objects of intellection, i.e. universals. This would mean that we cannot have knowledge of a particular sensible object as such. But it would not necessarily mean that sensation makes no contribution to knowledge. One might hold, with Aristotle, that sensation somehow makes it possible for us to learn new intelligibles or universals, even though intellection and knowledge grasp universals, not sense-particulars. A second, stronger thesis would be to say that sensation not only has different objects from intellection, but also makes no contribution to intellections. According to this thesis there would be an ‘epistemic gap’ between sensation and intellection, with intellection operating on its own, and sensation serving only to distract us from the good function of intellect.61 Adamson argues that the opening paragraph of the second chapter of the Treatise may support the weak thesis that sensation and intellection have different objects but that sensation may contribute to intellection. However, it does not support the strong thesis that sensation cannot contribute to intellection: ‘If I grasp the universal man … what I am doing is grasping a truth that applies to all particular men. In that case, it seems plausible that we learn the universal precisely through our experiences with various particulars that fall under that universal’.62 Al-Kindī asserts this position in his treatise ‘On the Quantity of Aristotle’s Books’: … the primary substance—I mean the sensible—is known in terms of its predicates—for sensation does not grasp it directly but through the mediation of quantity and quality. Accordingly, the negation of quantity and quality means the negation of the knowledge of primary substance; indeed the true, fixed science of philosophy is the science of substances. Knowledge of secondary substance does not change because its objects do not change and is void of change and its flux. Secondary qualities are initially accessed (yutaṭaraq ilayhā) through knowledge of primary substances.63 However, Adamson argues that sensation is not as central as it may sound pointing to the context of this statement by al-Kindī. Adamson explains that al-Kindī ‘ … is defending the place of the mathematical, propaedeutic sciences as a pre-requisite for all other knowledge. And these mathematical, sciences (arithmetic, geometry, astronomy and harmonics) are not studies of sensible particulars, but of numbers, lines, rations, and the like. Sensation is only required for these sciences insofar as we first encounter number, line, ratio and so on as inhering in a given sensible object’.64 In other words sense perception is only the medium through which we encounter the qualitative and quantitative attributes with which mathematics deals in abstraction from the plenum of the sensible particulars. According to Adamson this does not imply that sensation contributes to knowledge. To substantiate his interpretation Adamson quotes the part of the paragraph that immediately follows: Sensory knowledge is the knowledge of primary substance, and is in flux due to the uninterrupted flux of what is known (this ends only when the [object of knowledge] itself ends, which means that it is wholly destroyed in its substance), or because of the multiplicity of sensible substance and the multiplicity of number. For if all that is numbered is finite, and it is always possible to multiply any number, then what is numbered is potentially infinite in magnitude—if not in the number of the individuals, then in the number of multiples. But what is infinite cannot be comprehended by any knowledge … 65 Adamson takes this passage to reinforce his claim concerning al-Kindī’s commitment to the weak thesis that sensation may only make possible the knowledge of new universals. Subsequently, Adamson reverts to al-Kindī’s argument in the opening passage of the second chapter of the Treatise regarding sensible particulars. All the sensible particulars are in a state of flux with regard to quantity and quality, which, in turn, are the basis of any possible form of knowledge, as indicated above. Adamson adduces al-Kindī’s claim that sensible particulars can comprise opposites; for instance a sensible particular can be both equal and unequal at the same time (for instance a stick can be equal in length to one stick and not equal in length to another). He suggests that al-Kindī’s position be read alongside Plato’s similar claim in the Phaedo. Hence, on the relation between sensation and rational cognition, he writes: ‘Sensation merely prompts the soul to uncover the intelligibles that are already within it, by reminding it of those intelligibles. The intelligibles cannot be somehow gleaned afresh from sensation’.66 Adamson uses al-Kindī’s discussion of the knowledge as recollection of universal intelligibles apart from any material differentiation in his treatise ‘On Recollection’ to suggest that al-Kindī may have very well endorsed the strong thesis, viz. that sensation does not have any epistemological function. Thus, Adamson argues that, in a number of his writings, al-Kindī leaves an ‘epistemic gap’ between sensation and intellection indicating that ‘he seems unaware or unconvinced by Aristotle’s strong empiricist commitments … ’67 In other words al-Kindī was not convinced that universals could be abstracted from sensible particulars. Adamson cites a passage from al-Kindī’s treatise On Rays, which challenges his claims about the Platonism of al-Kindī. In this passage, al-Kindī suggests that reason uses sensation to grasp the common forms, the universals shared by sensible particulars. Adamson concedes that his Platonic conception of the ‘epistemic gap’ between sensation and intellection in al-Kindī’s epistemology could be questioned. However, he acknowledges this only as ‘the exception that proves the rule’ and on the assumption that On Rays ‘was written late in al-Kindī’s career’.68 However, the sharp contrast al-Kindī draws between sensation and intellection has complex methodological and epistemological connotations that extend beyond the challenge to Adamson’s Platonic description of its origins. Methodologically, this contrast follows al-Kindī’s assertions in the first section of the Treatise that the different domains of science that inquire into different regions of beings, in turn accessed through different faculties, have to be separated. Hence, al-Kindī advises that inquiries into non-sensible substances must employ purely rational arguments; the science that encompasses all such inquiries is metaphysics. Similarly, mathematics must not be used in physics proper since physics is the study of sensible substances qua motion or change.69 Different regions of beings that are accessed by different faculties are to be approached through different domains of science. Epistemologically, the ramifications of such a contrast are rather ambiguous. As shown above, Adamson demonstrated that al-Kindī’s characterization of the relation between sensation and intellection is not always consistent. However, there is an important strand to this complex picture worth noting. While it is indisputable that al-Kindī would consider knowledge proper to be pure thinking of universals devoid of any sensible qualities, al-Kindī seems to be consistent in asserting that both sensation and intellection are modes of finding out that complement each other epistemologically. In his treatise titled ‘On Dreams’, al-Kindī writes ‘ … therefore Plato said that the soul is the receptacle of all things sensible and intelligible. Hence, the soul knows by nature. For all knowledge comes from reason or the senses … ’.70 This sheds new light on the relation between sensation and intellection. Beyond the weak and strong theses suggested by Adamson to define the relation between sensation and intellection, it seems that al-Kindī is more interested in identifying two types of perception that equally contribute to our knowledge of the world even if knowledge proper is that of rational universals. In other words, al-Kindī is interested in defining the two domains of perception as two modes corresponding to two faculties through which the soul can know or at least relate to a knowable thing. The same thing can be approached differently from the respective perspectives of these two domains. This difference is not due to a change in the essence of this same thing; rather, it depends on the mode of finding out applied to it, which in turn determines what of its characteristics it will disclose. The interpretation of al-Kindī’s argument for the finitude of time suggested in the first part of this study supports this interpretation of the contrast between intellection and sensation. Shifting the focus of concern from the object to the soul that perceives the object enables al-Kindī to respond to the problematic implications of the thesis that the existence of time depends on the soul. Time qua intellectual perception can be thought of mathematically as a continuous quantity. Further, metaphysically speculative claims may be made about time as such in connection with essentially rational concepts/problems, i.e., concepts/problems that are not part of nature. Time qua sensation is actual, divided time that is inseparably connected with motion and sensible substances, the immediate object of sensation. This is the advantage that al-Kindī’s position has over that of Philoponus on the one hand, and Alexander and Aristotle, on the other. The former critiqued the theory of the eternity of the world by reconstructing the concept of motion, restricting himself to, essentially, physical parameters and avoiding all theoretical speculation. Al-Kindī develops this position by showing the essential role of the soul as a necessary condition for physical inquiry. While Alexander and Aristotle pointed out the constitutive role of the soul, al-Kindī’s insistence on separating the different domains of theoretical sciences as corresponding to different modes through which the soul comports itself towards the same thing enables him to overcome the difficulties entailed in this position. To reinforce my interpretation, I will make one last suggestion about the epistemic nature of the moral distinction between perceptual being and rational/ideal being in al-Kindī’s system. II.2 Between al-Kindī and Avicenna: the logical vs. epistemic modality of being and the relation between sensation and intellection Adamson has argued that al-Kindī uses two distinct yet reconcilable notions of being.71 The first is simple being. Simple being is the being of God to whom unity is exclusively attributed. In this vein, simple being is being devoid of any essence or attributes. Hence, simple being in itself does not allow for predication. Simplicity, here, consists in that being cannot be qualified. In this sense simple being is different from unlimited being. Adamson explains that this conception of simple being is distinct from the concept of unlimited being endorsed by the Muʿtazili school of rational theology. According to this school, God is unlimited because He is identical with His attributes/predicates. In contrast, Adamson argues that, for al-Kindī, God is simple being in itself because nothing can be predicated of Him. Therefore Adamson likens al-Kindī’s sense of simple being to Aristotle’s sense of prime matter: ‘ … being’ [that is simple being] is treated as analogous to Aristotelian matter … [Aristotle] … describes matter as the ultimate subject of predications that underlies all features of a thing. Also like Aristotelian matter, being subsists through change as becomes clear … when al-Kindī says that being ‘does not change’. The point is an intelligible one: even in the case of substantial corruption (such as death in the case of a human), there is not an absolute destruction of being but merely of the way the thing is. This is why the corpse that remains when the human is no longer alive is yet something that exists. Finally, like Aristotelian matter, mere being must be simple, where ‘simple’ again means without predicates. For as the ultimate subject of predication, being itself cannot be further analyzed into a complex of subject and predicates.72 By contrast to simple being, complex being is the being of the world, everything that is generated. Adamson paraphrases al-Kindī on complex being as follows: ‘ … the body of the universe causes it to have a limited, temporal existence. Here it is more natural to understand anniyya as “nature” or “essence”. Indeed, at one point he makes a remark that equates huwiyya, “being” with mā huwa “what a thing is” ’.73 Adamson summarizes al-Kindī’s general position on being as follows: When he [al-Kindī] speaks of ‘being alone’, he means the mere act of existing that is prior to, and the subject of, the existent’s essence and other predicates. But he also speaks of ‘a being’, by which he means a fully constituted being that is already considered to have an essence. On this latter notion, the being of each thing will be distinct from the being of anything else; on the former notion, being is mere existence and belongs to anything that God has seen fit to create.74 Adamson argues that al-Kindī’s concept of simple being anticipates some aspects of Avicenna’s metaphysics associated with the distinction between essence and existence. Both Avicenna and al-Kindī maintained that being is distinct from attributes, whether essential or not. For instance, existence for Avicenna can be thought in complete independence of essence. Similarly, God for both Avicenna and al-Kindī is pure existence, which, in turn, instantiates the complex existence of the essence of every other being. Adamson contends, however, that al-Kindī’s concept of complex being does not show any aspects that could possibly anticipate Avicenna’s distinction between essence and existence. For al-Kindī the existence of object X consists in satisfying the essential features of X. If X is a human being, the essential features would be animality and rationality. Similarly, al-Kindī and Avicenna cannot be reconciled on the priority of essence over existence: al-Kindī is committed to the priority of existence—the reverse of Avicenna’s position.75 I would like to suggest another parallel between al-Kindī and Avicenna, specifically concerning the relationship between the essence and existence of complex being. Olga Lizzini has argued that what is most distinctive about Avicenna’s metaphysics is its modal orientation.76 She suggests that what is most provocative about Avicenna’s notion of being is not whether or not essence precedes existence, as many scholars have tended to remark. Rather, according to Lizzini, what matters most is whether the existent under consideration is necessary or possible, as Avicenna argues in book I.6 of the Ilāhiyyāt of his al-Shifāʾ. In other words, Avicenna gives priority to the logical determination of the nature of the essence and whether or not it is necessary rather than possible over and above whether or not it is existentially instantiated. A similar modal orientation can be discerned in al-Kindī’s epistemology. He identifies two modes of finding out: intellection and sensation. These two modes identify two modalities through which any substance can be epistemically accessed. In turn, these two modalities allow for different senses of being of the same substance. As argued above, the being of time mathematically considered can be potentially infinite. However, its being is finite in actuality. On a mathematical level it is a continuous quantity that is infinitely divisible. In actuality, it is associated with the ‘being’ (huwiyya) of a particular sensible substance as grasped by a perceiving soul. Since any sensible substance is constantly in motion and since time is the quantity measuring this motion, then time must be finite. This is another way of thinking about the relation between simple and complex being. Simple being in relation to every substance defines the modality in which it is conceived and presented apart from its essence and attributes. Hence, it is epistemic modality that determines the nature of being of the essence, and not vice versa. Unlike Avicenna’s strictly logical conception of the modalities of essences, I am suggesting that al-Kindī associates modality with the faculties of the soul. Following the Aristotelian line of thinking in the De Anima, al-Kindī associates each faculty of the soul with a specific domain of beings. However, in contrast with Aristotle (and definitely more lucidly), al-Kindī argues that the being of certain classes of beings can be understood differently based on the way they are considered by different faculties of the soul. In addition, there is a further connection between the reception by different faculties of the soul of the same object and the constitution of the domains of sciences in al-Kindī’s epistemology. In the domain of physics senses and empirical observation play a definitive role in the constitution of judgments about nature. By contrast, the domain of metaphysics essentially relies on logical investigations of universals and relations among them just like the domain of mathematics, in its own right, relies on abstract quantitative (kammī) analysis. The results of the investigations in each scientific domain cannot be extended to a different domain because their principles are different. Notwithstanding the distinction I have just expounded, between al-Kindī’s epistemic modality and Avicenna’s insistence on the constitutive role of logical modality, I would like to argue that both are inspired by a similar critique of the primacy of metaphysics in classical antiquity. Let me begin with this important observation by Charles Kahn in ‘Being and Existence’: … My general view of the historical development is that existence in the modern sense becomes a central concept in philosophy only in the period when Greek ontology is radically revised in the light of a metaphysics of creation; that is to say under the influence of biblical religions. As far as I can see, this development did not take place with Augustine or with the Greek Church Fathers, who remained under the sway of classical ontology. The new metaphysics seems to have taken shape in Islamic philosophy in the form of a radical distinction between necessary and contingent existence: between the existence of God on the one hand and that of the created world on the other. The old Platonic contrast between Being and Becoming, between the eternal and the perishable (in Aristotelian terms between the necessary and the contingent) now gets reformulated in such a way that for the contingent being of the created world (which was originally present only as a possibility in the divine mind) the property real existence emerges as a new attribute or ‘accident,’ a kind of added benefit bestowed by God upon possible beings in the act of creation. What is new is the notion of radical contingency: not simply the old Aristotelian idea that many things might be other than they in fact are—that many events might turn out otherwise—but that the whole world of nature might not have been created at all: that it might not have existed. I leave it to the historians of Islamic and medieval philosophy to decide how far my hypothesis is correct and to determine just when, or in what stages, the new concept of existence was formulated … 77 With that observation in mind, it will be worthwhile to indicate how my interpretation of the epistemic gap between sensation and intellection in al-Kindī’s theory of knowledge could clarify how his critique of Aristotle’s theory of the eternity of the world contributed to the reconceptualization of the question of being in Islamic philosophy—no less than did Avicenna’s logical distinction between necessary and possible essences did. Al-Kindī does not definitely relegate the problem of existence in the way that Avicenna does through his distinction between essence and existence and his emphasis on the fundamentality of the logical modality of essence. However, al-Kindi’s distinction between sensation and intellection as two possible epistemic modes enables a reconceptualization of the relation between epistemology and metaphysics in a manner that was inconceivable by Greek ontology. If the interpretation of the epistemic gap suggested above is correct, then al-Kindī’s position on complex being can be revisited as follows. Just like ‘time’, the being of any complex being depends on the epistemic mode through which it is being considered. From a purely rational/metaphysical perspective that being would be defined in terms of the existence of its essential attributes; from a physical perspective it would be the huwiyya or how these essential attributes are constituted for a perceiving soul. In this regard, al-Kindī may be anticipating the later distinction in Islamic philosophy and philosophical theology between mental existence (al-wujūd al-dhihnī) and perceptual external existence (al-wujūd al-khārijī). This was later formalized in the twelfth century under the influence of al-Ghazālī (d. 505/1111) and Fakhr al-Dīn al-Rāzī (d. 606/1209). What is more certain is that al-Kindī’s epistemic gap reflects or at least makes possible a reformulation of the ontological commitment of Greek naturalism. Embedded in this reformulation is a favouring of epistemic conditions over the concern with metaphysical claims. If Charles Kahn’s interpretation of the significance of Avicenna’s distinction between essence and existence and the necessary and the contingent in terms of re-questioning Greek ontology are correct, I would argue that al-Kindī epistemic modality contributes to the same objective. Conclusion In this paper I have pursued two objectives. The first was to clarify the importance of al-Kindī’s peculiarly formulated second counter proof of the eternity of the world. Contrary to Davidson’s thesis, and its adoption and reformation by Adamson, I showed that this step is not superfluous. Rather, it leads to an important aspect of al-Kindī’s thought, which reconstructs and creatively develops, a classical Aristotelian position that was espoused in the tradition of Aristotelian commentators by Alexander in On Time. The conclusion drawn about al-Kindī’s endorsement of the relation between time and the soul indicates that, while al-Kindī was influenced by Philoponus’s critiques of Aristotle, he sought to find a new ground for the latter’s interest in restricting physical arguments to actual, finite experience. My second objective was to reinforce this suggested reading of the second proof by indicating its connection with al-Kindī’s commitment to the demarcation of the boundaries of theoretical sciences. I argued that this demarcation hinges on the correspondence between these sciences and the two different modes through which the soul comports to objects: sensation and intellection. If this interpretation is correct, this helps mitigate key problematic implications, as Sharples indicated, of Alexander’s and Aristotle’s position on the dependence of time on the soul. Further, the findings of the second section sought to clarify the ‘epistemic gap’ between intellection and sensation pointed out by Alfred Ivry and further developed by Adamson. I have tried to situate this argument in the broader context of al-Kindī’s epistemological commitments, explaining how it is in harmony with the findings of the first part of the paper, and exploring briefly some affinities it shares with Avicenna’s modal interests. Footnotes 1 Al-Kindī advanced three major counter-arguments to refute Aristotle’s and the Neo-Platonist’s position on the eternity of the world. In this paper I focus on the second one. The order of the arguments I follow is that of Herbert Davidson in his landmark article: ‘John Philoponus as a Source of Medieval Islamic and Jewish Proofs of Creation’, Journal of the American Oriental Society, 89/2 (1969): 357–91. 2 All the translations of Aristotle’s works cited in this paper are quoted from Jonathan Barnes, The Complete Works of Aristotle (Princeton, NJ: Princeton University Press, 1995), with references in accordance with the Bekker/Prussian Academy of Sciences standard pagination. The Physics translation is by R. P. Hardie and R. K. Gaye. 3 Herbert Davidson, ‘John Philoponus as a Source’, 357–91, at 370–5. 4 Peter Adamson, Al-Kindī (New York: Oxford University Press, 2007), 74–105. See also Peter Adamson and Peter E. Pormann, The Philosophical Works of Al-Kindī (Karachi: Oxford University Press, 2012), 3–81. 5 Brief references will also be made to Philoponus’s other major critique, in Contra Proclum, of Proclus’s appropriation—in On the Eternity of the World and his lengthy commentary on the Timaeus—of Aristotle’s argument for the eternity of the world. 6 See section I.1 below for references and some discussion of the work of Sorabji and Sharples. 7 While the existential connection between time and the soul was acknowledged in the works of a number of ancient commentators including Simplicius, I take note of Alexander of Aphrodisias’s treatise On Time because of its importance and its having been translated around the time of al-Kindī. The treatise was translated by Ḥunayn as is stated on the Arabic manuscript discovered and published by ʿAbd al-Raḥmān Badawī in 1971 (see n.10 below). It was mentioned in Abū al-Rayḥān al-Bīrūnī (d. 439/1048), Kitāb al-Bīrūnī fī taḥqīq mā li-l-Hind min maqūla maqbūla fī al-ʿaql aw mardhūla (Hyderabad: Dāʾirat al-Maʿārif al-ʿUthmāniyya, revised edn., 1958), 271–6. It is difficult to determine historically when Ḥunayn ibn Isḥāq did the translation and whether it could have been available to al-Kindī when he most likely composed his Treatise during the reign of the Caliph al-Muʿtaṣim (r. 218–27/833–42). However, given the overlap between Ḥunayn’s and al-Kindī’s time it is not unreasonable to venture that the latter was somehow familiar with it. Nonetheless, Alexander’s position on the nature of time is invoked here as a philosophical forerunner to the position that, as this study suggests, al-Kindī takes in his second proof. The reference to Alexander’s On Time is not intended as a historical claim that al-Kindī actually read it, and the validity of the argument here is not dependent on the validity of such a claim. 8 See R. P. Hardie and R. K. Gaye, The Complete Works of Aristotle, i. 223a16–28. 9 Richard Sorabji, The Philosophy of the Commentators 200–600 A.D. (Ithaca, NY: Cornell University Press, 3 vols., 2005), ii. 203. 10 Ibid, 201–2 (parenthesis in the original). Richard Sorabji translates this passage directly from the Arabic translation of On Time by Ḥunayn ibn Isḥāq, published in ʿAbd al-Raḥmān Badawī, Shurūḥ ʿalā ArisṠū mafqūda fī al-Yunāniyya wa-rasāʾil ukhrā (Beirut: Dār al-Mashriq, 1971), 20–4. The Latin translation of the Arabic text by Grerad of Cremona in the twelfth century ce was translated by R. W. Sharples, ‘Alexander of Aphrodisias, On Time’, Phronesis, 27/1 (1982): 58–81. Acknowledging that this part is a commentary on Aristotle’s Physics 223a16–29, Sharples explains (p. 70) that ‘… both Aristotle and Alexander hold that time would not exist in the absence of soul. Alexander however, has two arguments for this conclusion where Aristotle has only one, explicitly. Both authors argue that time depends on soul on the ground that it is soul that numbers movement; but Alexander adds, apparently, as a separate argument, that in the absence of the soul there would not be any movement. It is his standard doctrine that the movement of the heavenly spheres is caused by the desire of their souls to emulate the changelessness of the Unmoved Mover. 11 Sorabji, The Philosophy of the Commentators, ii. 202. 12 Ibid. 13 Ibid. 14 Sharples, ‘Alexander of Aphrodisias, On Time’, 71. 15 Philoponus, Against Aristotle on the Eternity of the World (transl. Christian Wildberg; Ithaca, NY: Cornell University Press, 1987), 18–31. 16 Philoponus’s interest in refuting the eternity of heavenly bodies and motion may be regarded as a possible response to the Aristotelian paradox that Sharples pointed out. The reason is that the refutation of the argument for the eternity of heavenly bodies is an implicit refutation of the possibility of eternal time or time ‘in itself’. 17 Ibid, 41–57. 18 Ibid, 58–68. 19 Ibid, 69–76. 20 Ibid, 77–91. 21 Ibid, 93–121. 22 The argument against pre-existing matter similarly constitutes the kernel of Philoponus’s arguments in Contra Proclum against Proclus’s On the Eternity of the World. Proclus resorted to the Aristotelian reductio ad infinitum proof of pre-existing matter in order to prove the eternity of the world and refute the Christian arguments for the creation ex nihilo. See in particular Books 9 and 11 of Contra Proclum in John Philoponus, Against Proclus On the Eternity of the World 9–11 (transl. Michael Share; London: Duckworth Publishers, 2010). See also: Frans A. J. De Haas, John Philoponus’s New Definition of Prime Matter (Leiden: Brill, 1997), 1–45. 23 Philoponus, Against Aristotle on the Eternity of the World, 123–4. 24 Ibid, 124–5. 25 Generation is a form of motion, as Philoponus acknowledges. However, in this context he is primarily concerned with locomotion as the most fundamental form of motion as distinct from change in quality (alteration), change in quantity (growth/diminution), or change in substance (generation and corruption). He thus agrees with Aristotle on the primacy of motions. For example, see Aristotle’s Physics 260b29–261a13. 26 Philoponus, Against Aristotle on the Eternity of the World, 125–7. 27 Ibid, 127. 28 See chs. 2–6 in De Haas, John Philoponus’s New Definition of Prime Matter. 29 Ibid, 285–90. 30 See fragments 114–16 in Philoponus, Against Aristotle on the Eternity of the World, 128–31. 31 See fragments 121–2, in ibid, 133–7. See also al-Ghazālī’s reconstruction of this argument in the ‘First Discussion of the Pre-Eternity of the World’ in Incoherence of the Philosophers. Al-Ghazālī, in contrast to Philoponus, associates erring with the limitation of the human estimative faculty and imagination which mislead the intellect. See Abū Ḥāmid al-Ghazālī, Tahāfut al-falāsifa (ed. Sulaymān Dunyā; Cairo: Dār al-Maʿārif, 2000), 110–14. 32 See fragments 123–4 in Philoponus Against Aristotle on the Eternity of the World, 137–8. 33 See fragments 125–6 in ibid, 139–40. 34 See fragment 132 in ibid, 143–6. 35 See al-Kindī, Rasāʾil al-Kindī al-falsafiyya (ed. Muhammad ʿAbd al-Hādī Abū Rīda; Cairo, Dār al-Fikr al-ʿArabi, 2 vols., 1950–3), i. 98–106. 36 See al-Kindī, Rasāʾil al-Kindī al-falsafiyya, 193–4. 37 Ibid, 185–92. 38 Ibid, 199–207. 39 See al-Kindī, Rasāʾil al-Kindī al-falsafiyya, i. 117. 40 Herbert Davidson, ‘John Philoponus as a Source’, 372. 41 al-Fārābī, Kitāb al-Ḥurūf (ed. Muḥsin Mahdī; Beirut: Dār al-Mashriq, 2nd edn., 1990), 110–28. 42 Ikhwān al-Ṣafāʾ, Rasāʾil Ikhwān al-Ṣafāʾ wa-khillān al-wafāʾ (ed. Khayr al-Dīn al-Ziriklī; Cairo: al-Maṭbaʿa al-ʿArabiyya, 4 vols., 1928), i. 350. (My translation.) 43 al-Fārābī, Kitāb al-Fuṣūṣ (Hyderabad: Dāʾirat al-Maʿārif al-ʿUthmāniyyā, 1346/1927), 2. 44 See al-Kindī, Rasāʾil al-Kindī al-falsafiyya, i. 21. (My translation.) To sum up: the huwiyya of an object ‘x’ refers to its self-sameness or identity in perception; it is the principle of its individuation inasmuch as it is perceived by a perceiving soul. As such, huwiyya has an epistemic import. In contrast, māhiyya refers to the Aristotelian species-form, in abstraction from any individual characteristics; and mawjūd is a general term of an essentially metaphysical import referring to anything of which existence is predicated whether mental existence (wujūd dhihnī) or external existence (wujūd khārijī). 45 Ibid, 119. (My translation.) 46 al-Kindī, Rasāʾil al-Kindī al-falsafiyya, i. 167. 47 Adamson, Al-Kindī, 84–5. 48 Alfred Ivry, Al-Kindī’s Metaphysics: a translation of Yaʿqūb ibn Isḥāq al-Kindī’s treatise On first philosophy’ (Fī al-Falsafah al-ūlā) (Albany, NY: State University of New York Press, 1974), 26. 49 Jonathan Lear, Aristotle: the Desire to Understand (Chicago, IL: University of Chicago Press, 1999), 74. 50 R. P. Hardie and R. K. Gaye, The Complete Works of Aristotle, i. 251a24–7. 51 Jonathan Lear, Aristotle: the Desire to Understand, 82–3. 52 Alfred Ivry and Peter Adamson both translate al-wujūd as perception; the use of the word perception is rather misleading given that it is conventionally used in Western epistemology to refer to sense perception. As Abū Rīda explains in the introduction to his edition of the Rasāʾil, it should be translated as finding inasmuch as it derives from the Arabic verb to find (wajada). This suggestion is supported by al-Fārābī’s account of the genealogy of the use of the term in Arabic in general and in theoretical sciences in particular. Al-Fārābī argues that the general Arabic term refers to what a human being finds existing (kāʾin) in another being/human being. However, it can refer to being as such whenever it is used without qualification in theoretical sciences to translate the Greek (ɛ͗στιν)—to be. Hence al-Fārābī argues that mawjūd has to be translated as being in an objective sense independent of its reception by any human being. But since al-Kindī is qualifying wujūd in this context by ‘human’, it is more appropriate to translate it as finding rather than perception. See al-Fārābī, Kitāb al-Ḥurūf, 110–28. It is noteworthy that al-Kindī makes an innovation that is completely overlooked by Ivry and Adamson in their commentaries on this passage. Isḥāq ibn Ḥunayn in his translation of Physics 184a17–184b13 uses the word al-umūr meaning matters/things to refer to the things ‘which are more knowable and clear to us and proceed towards those which are clearer and more knowable by nature … ’. See Arisṭūṭālīs, al-Ṭabīʿa (ed. ʿAbd al-Raḥmān Badawī; Cairo: al-Dār al-Qawmiyya li-l-Ṭibāʿa wa-l-Nashr, 2 vols., 1964–5), i. 3. Abū Bishr Mattā uses the same Arabic word al-umūr to translate a similar passage in the Posterior Analytics I.2, 72a 1–5. See Arisṭūṭālīs, Manṭiq Arisṭū (ed. ʿAbd al-Raḥmān Badawī; Cairo: Dār al-Kutub al-Miṣriyya, 3 vols., 1948–52), iii. 314. In light of the results drawn from the last section, the importance of the use of wujūd as against umūr is that al-Kindī redefines the distinction between two modes of finding instead of two species of objects. However, in keeping with the terminology mostly employed in medieval philosophy, I will render wujūd as perception. 53 Ivry, Al-Kindi’s Metaphysics, 61. 54 See Aristotle, Physics, 184a17–184b13. 55 al-Kindī, Rasāʾil al-Kindī al-falsafiyya, i. 107. 56 Ibid, 107–8. 57 Ibid, 108–9. 58 Ibid, 167, 295. 59 Ivry, Al-Kindi’s Metaphysics, 135–6. 60 al-Kindī, Rasāʾil al-Kindī al-falsafiyya, i. 312–62. 61 Adamson, Al-Kindī, 127–8. 62 Ibid, 129. 63 al-Kindī, Rasāʾil al-Kindī al-falsafiyya, i. 372. (My translation.) 64 Adamson, Al-Kindī, 129. 65 Ibid. 66 Ibid, 133. 67 Ibid, 134. 68 Ibid. 69 See al-Kindī, Rasāʾil al-Kindī al-falsafiyya, i. 110–12. 70 Ibid, 301. (My translation.) 71 Peter Adamson, ‘Before Essence and Existence: al-Kindī’s Conception of Being’, Journal of the History of Philosophy, 40/3 (2002): 297–312. 72 Ibid, 302. 73 Ibid, 307. 74 Ibid, 308. 75 Ibid, 310. 76 Olga Lizzini, ‘Wuǧud/Mawǧud/Existence–Existent in Avicenna: A key ontological notion of Arabic philosophy’, Quaestio, 3 (2003): 111–38. 77 See Charles Kahn. Essays on Being. (New York: Oxford University Press, 2009), pp. 62–4. © The Author(s) (2018). Published by Oxford University Press on behalf of the Oxford Centre for Islamic Studies. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Journal of Islamic StudiesOxford University Press

Published: Mar 29, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off