Chapter Seven: The Interaction of Philosophy and Dogma

I The Eclipse of Theological Rationalism

As we mentioned earlier, the rise of Scholastic theology in the middle of the eighth century was the outcome of a new spirit of inquiry, which the introduction of Greek philosophy in the Muslim world had sparked. In some cases, however, the interaction of philosophy and dogma resulted in a gradual cleavage between the two. The systematic philosophers, like al-Farabi and Ibn Sina, tried hard to lessen the effect of such cleavage by emphasizing the areas of agreement and the common concerns of philosophy and dogma. Some, such as al-Kindi, went so far as to espouse the cause of dogma almost unconditionally and sought to erect a compact intellectual edifice on the foundation of dogma.

A gradual reaction to rationalism in theology, championed originally by the Mu‘tazilah, was to set in less than a century after the death of the founder of that school, Wasil b. ‘Ata'. We have already discussed the role which the great theologian and jurist Ahmad b. Hanbal, as well as the ‘Abbasid caliph al-Mutawakkil, played in the reversal of the pro-Mu‘tazilite policies of al-Ma'mun in the middle of the ninth century. However, the theological influence of the Mu'tazilah did not cease altogether as a result of al-Mutawakkil's policy of repression. Despite the virtual triumph of the Hanbali and Traditionist party, the spirit of theological inquiry was not completely snuffed out. In its pure form, the primitive traditionalism of the early jurists and exegetes was gone forever. The new Traditionism or orthodoxy was a qualified one that stemmed from the Mu‘tazilite movement itself. Its rise is associated with the name of Abu'l Hasan al-Ash‘ari (d.935), who, according to the traditional account, studied theology with al-Juba'i, head of the Basra branch of the Mu‘tazilite school but broke away from that school at the age of forty. The Prophet appeared to him in a dream and urged him to "take charge" of the Muslim community, whereupon al-Ash‘ari ascended the pulpit at the mosque of Basra and proclaimed his recantation and his determination to make public "the scandals and follies" of the Mu‘tazilah.

A debate with his master, al-juba'i, concerning God's justice and man's worthiness brings out vividly his original anti-Mu‘tazilite sympathies. Whether historical or not, this debate is significant in so far as it illustrates one of the cardinal issues on which al-Ash‘ari broke with the Mu‘tazilah. The pupil asks his master: What will be the fate in the after-life of three brothers, one of whom dies in a state of grace, one in a state of sin, and one in a state of innocence (i.e., before he comes of age)? The righteous brother, answers al-juba'i, will be consigned to paradise, the sinner to hell, and the third to an intermediate position. Al-Ash‘ari then asks: What if the third brother were to ask to be allowed to join his more fortunate brother? This privilege, replies al-juba'i, would be denied him on the ground that the first brother was admitted to paradise on the strength of his good works. If the third brother were to protest that if he had been given a long life he would have lived righteously, God would have replied: I foresaw that you would not and therefore chose to spare you everlasting damnation in hell. At this, the brother Who had died in sin exclaims: Surely, Lord, you foresaw my own plight, as well. Why, then, did you not deal with me as mercifully as you have dealt with my other brother?

We are told that al-juba'i was unable to say what God's possible answer to such protestations might be, on the Mu‘tazilite assumption of the unqualified justice of God. The corollaries drawn by al-Ash‘ari constitute the substance of his view of God's absolute omnipotence and sovereignty in the world and the finality of his moral and religious decrees. These decrees are entirely independent of any conditions, moral or other, apart from God's absolute fiat. To Him it belongs to order human life as He pleases, and to the "servant" to obey without question. Contrary to the contention of the Mu‘tazilah, the human agent plays no part in the drama of choosing or doing and reaps none of the moral or religious fruits accruing from such initiative. In their desire to stress man's moral freedom and responsibility, the Mu‘tazilah had described him, somewhat extravagantly, as "the creator of his deeds." To al-Ash‘ari, such blasphemous language was tantamount to the denial of God's uniqueness as the sole Creator and Sovereign of the world, and consequently implied the recognition of two creators, in the manner of the Manichaeans (Majus).

The vindication of God's absolute power and sovereignty in the world had certain moral implications, which al-Ash‘ari was quick to draw. To deny man's role in the drama of moral action and decision and to impute the responsibility for his deeds and volitions to God involved the repudiation of God's justice. However, the claim that man's deeds are the result of God's "decree and preordination" did not necessarily imply, according to him, the nullification of His justice. Injustice can only denote the transgression of what has been prescribed by a superior, or the perpetration of what falls outside the domain of the doer. In both cases, injustice cannot be imputed to God, Who is the undisputed master and lawgiver of the universe and Who owes no allegiance to anyone whatsoever.

On the question of the attributes of God and the creation of the Qur'an, the position of al-Ash‘ari was equally at variance with that of his Mu‘tazilite master, on the one hand, and that of the crude anthropomorphists or literalists, on the other. Moved by the desire to retain the Concept of the full-blooded Creator-God of the Qur'an, he opposed the Mu‘tazilite tendency to divest God of His positive attributes, and argued, according to a twelfth-century historiographer and fellow-Ash‘arite, al-Shahrastani, that the essential divine attributes of knowledge, power, and life are eternal and subsist in God's essence. They cannot, however, be said to be either identical with this essence, as the Mu‘tazilah claimed, or not identical with it. For this would mean that God's knowledge, power, or life is the same as God, So that one could address one's petitions to God's knowledge, power, or life instead of to God Himself, which is absurd.

The rationalization of the inherence of the attributes in God which the Mu‘tazilah attempted is not fully worked out by al-Ash'ari or his followers. How these attributes are to be distinguished from God's essence, in which they inhere and yet introduce no plurality into it, al-Ash'ari just refused to say. In this respect he is content to revert to the position of the early Traditionists, such as Malik b. Anas, who is reported to have argued, in the matter of God's "sitting upon the throne," that the "sitting is known, whereas its mode is unknown. Belief in its truth is a duty, and its questioning a heresy ."

In his polemical works, however, al-Ash'ari is as concerned to refute the views of the "negators of the attributes," i.e., the Mu‘tazilah, as he is to refute the position of the literalists and anthropomorphists. In their deference to Scripture, the latter had gone so far as to attribute corporeity to God, chiefly on the grounds that the text of the Qur'an undeniably stipulated it. Thus Qur'an 75:22-23 speak of the ability of the faithful to perceive God on the Last Day, and 7:54, and 20:5 speak of His sitting upon the throne. The anthropomorphists, such as Hisham b. al-Hakam, ‘Abdullah b. Karram, and their followers in the ninth century, had not hesitated to draw from such Qur'anic passages their full logical consequences and to conceive of God, as Ibn Rushd will say later, simply as an "eternal man" endowed with gross corporeal qualities.

The use of logical argument in matters of theology , and its permissibility, Should first be justified satisfactorily, however. Al-Ash‘ari's position, though reactionary by the standards of the philosophers and thoroughgoing rationalists, is certainly nuancé. Against the literalists and Traditionists, who questioned the permissibility of deduction or analogy, al-Ash'ari invokes the authority of the Qur'an, which recognizes the principle of analogy and employs it effectively in numerous passages. In a tract devoted to the systematic discussion of this question and entitled Vindication of the Use of Theological Proof (Kalam), this ex-Mu‘tazilite doctor's anti-Traditionist views on an issue which split the ranks of tenth-century theologians are clearly exhibited. The use of analogy, as indeed the whole method of dialectic or deduction, is repudiated by the Traditionists on the ground that the Prophet, who had dealt with every aspect of religion or morals essential to salvation, has not touched on the question of dialectic (Kalam) at all. Hence recourse to it constitutes an heretical departure (bid‘ah) from what is traditionally and authoritatively received.

This argument from silence is artfully turned by al-Ash‘ari against the Traditionists, who, by the same token, are just as heretical themselves, since their claim has no basis in the pronouncements or sayings of the Prophet either. More important still is the fact that the Prophet was fully conversant with the questions of motion and rest, accident and body, divine attributes, and so on, with which theology is concerned. However, they are referred to in the Traditions and the Qur'an in general terms only, and it is on such references that the whole of theology is based.

Finally, the silence of the Qur'an and the Traditions on those questions that were subsequently dealt with by the theologians or the jurists is easily justified. The Muslim community was not faced with the difficulties or doubts which eventually led to them, or else the Prophet would have laid down explicitly the principles for solving them. As a result, the jurists and theologians in attempting to solve them had no other recourse than to draw analogies with what was explicitly laid down in Scripture. For it is the duty of every "reasonable Muslim" in such matters, al-Ash‘ari argues, "to refer them to the body of principles consecrated by reason, sense-experience, and common sense."

In applying this qualified rationalism to the cardinal questions debated in theological circles at the time, al-Ash‘ari, though in fundamental disagreement with the Mu‘tazilah, is nonetheless anxious to justify his opposition to them on rational grounds. The result is that his method is analogous to that of the Mu‘tazilah, whereas his doctrine is substantially a restatement of Traditionist or Hanbali theses.

If we take the Mu‘tazilite concept of free will as an instance, this dichotomy is clearly brought out. In the Ibanah, al-Ash‘ari describes the arbitrary power of God in terms that leave hardly any scope for human initiative:

We believe that Allah has created everything, by simply bidding it: Be, as He says [in Qur'an 16:42]: "Verily, when we will a thing, our only utterance is: ‘Be' and it is"; and that there is nothing good or evil on earth, except what Allah has preordained. We hold that everything is through Allah's will and that no one can do a thing before he actually does it, or do it without Allah's assistance, or escape Allah's knowledge. We hold that there is no Creator but Allah, and that the deeds of the creature are created and preordained by Allah, as He said [in Qur'an 37:94]: "He has created you and what you make" ...we hold that Allah helps the faithful to obey Him, favours them, is gracious to them, reforms and guides them; whereas He has led the unfaithful astray, did not guide or favor them with signs, as the impious heretics claim. However, were He to favor and reform them, they would have been righteous, and had He guided them they would have been rightly guided. ...But it was His will that they should be ungodly [singular: kafir], as He foresaw. Accordingly He abandoned them and sealed their hearts. We believe that good and evil are the outcome of Allah's decree and preordination [qada' wa qadar]: good or evil, sweet or bitter, and we know that what has missed us could not have hit us, or what has hit us could not have missed us, and that creatures are unable to profit or injure themselves, without Allah.

In this vindication of the omnipotence of God and the powerlessness of the creatures al-Ash‘ari simply reaffirms the Qur'anic Concept of the God-Despot, whose decrees are both irreversible and inscrutable. At the back of this polemic, however, is the view of the Mu‘tazilah that man is the "creator of his deeds," and consequently a fully free and responsible agent. The Concept of a co-creator with God, according to al-Ash‘ari, amounts to polytheism and involves a radical curtailing of God's absolute power. Despite these strictures, he does not concur with the Traditionists in their claim that man does not play any part whatsoever in the drama of moral activity. In his doctrine: of al-kasb, or acquisition of the merit or demerit for the deed done, al-Ash‘ari seeks a way out of the moral dilemma of responsibility, without sacrificing the omnipotence of God. Voluntary actions, in his view, are created by God, but acquired by the human agent or imputed to him. Creation differs from acquisition in that the former is the outcome of "eternal power," whereas the latter is the outcome of the "created power" of the agent, So that the same action is said to be created by the one and acquired by the other. Stated differently, man acquires the credit or discredit for the deed created by God, since it is impossible that God should acquire it in time, while He is its author eternally. In this subtle verbal distinction between what is acquired in time and what is created or predestined eternally, lies according to al-Ash‘ari, the distinction between voluntary and involuntary action, and also that between the merit or demerit which attaches to the latter. Man, as the locus or bearer of "acquired" action, becomes responsible for such action, whereas for involuntary action, such as trembling or falling, etc., he is totally irresponsible. The fundamental relation between the two forms of action, according to al-Ash‘ari and his followers, is that man is intuitively conscious of the difference between the one action and the other.Thus, rather than restore to man the freedom of which the extreme determinists (al-Jabriyah) had robbed him, al-Ash‘ari is content restore to him the consciousness of his subjection to the "eternal power." Through this subtle distinction, the predestinarian presuppositions of the Traditionists and determinists are not repudiated, but their linguistic sting is removed without surrendering the substance of the predestinarian thesis. The elaboration of this peculiar ethical position, as well as the occasionalist world-view on which it rested, should perhaps be left to a subsequent section, because of the part which the successors of al-Ash‘ari played in developing or refining it.

The historical significance of al-Ash‘ari's "reform" lies not in the elaborateness of his solutions of the theological problems raised by the Mu‘tazilah, but rather in his willingness to exploit their dialectical method, and, ipso facto, to moderate the claims of the Traditionists and antirationalists to whom he was temperamentally drawn. If his theological position, expressed in the classic formula of bila kaifa (ask not how) must be described as agnostic, it is nonetheless to be clearly distinguished from the blind agnosticism of the religious bigot who will entertain no questions whatsoever. For his was the qualified agnosticism of the earnest seeker who ends up by asserting, rightly or wrongly, the inability of reason to plumb the mystery of man in relation to God, or of God in relation to man.

II The Ash‘arite school and the Formulation of the Occasionalist Metaphysics of Atoms and accidents

The elaboration of the implication of al-Ash‘ari's new theological outlook was left chiefly to his successors in the tenth and the eleventh centuries. Apart from the substance of their anti-Mu‘tazilite creed, their attention was now centered on two fundamental questions: (1) the nature and limits of rational knowledge in relation to religious truth (‘aql AS. Sam‘), and (2) the metaphysical framework in which the concept of God's sovereignty and omnipotence should be expressed. Neither of these questions appears to have been discussed with any thoroughness by the founder of Ash‘arite movement himself.

The first major figure in the history of Ash‘arite school was Abu Bakr al-Baqalani (d. 1013), who belongs to the second generation of Ash‘arite doctors. This theologian, who is credited by later authors with refining the methods of Kalam, gives in his al-Tamhid the first systematic statement of the Ash‘arite doctrine and its metaphysical framework.

The book opens with a discussion of the nature of knowledge or science (‘ilm), in a manner which sets the pattern for similar Ash‘arite treatises such as al-Baghdadi's Usul al-Din and al-Juwayni's al-Irshad, but it has a distinctly modern ring. Thus, ‘ilm is defined by the author as "the knowledge of the object, as it really is." The object in question is then shown to include both that which is and that which is not (al-ma‘dum), which the Mu‘tazilah but not the Ash‘arites had declared to be a thing (shay'). Such ‘ilm falls into two major categories: the eternal knowledge of God and the temporal or created knowledge of creatures capable of cognition, such as men, angels, jinn, etc. The latter knowledge is subdivided in turn into necessary (or intuitive) and discursive.

Necessary knowledge is knowledge which can not be doubted. A subsidiary meaning, however, is that which cannot be dispensed with i.e. needful. Discursive knowledge, on the hand is knowledge which is the result of prolonged reflection, or, stated differently, knowledge which rests on necessary or empirical knowledge.

Such necessary knowledge is acquired through one or the other of the five senses and is essentially indubitable. However, there is a type of necessary knowledge which is not a matter of sensation, but is the result of the immediate apprehension of the mind, for instance man's knowledge of his own existence and his inner states or affections, such as pleasure or pain, love or hate, knowledge or ignorance. To this should also be added the knowledge of the truth or falsity of indicative statements, as well as the second-intention type of knowledge, such as the knowledge of what makes shame shameful, fear fearful, etc.

The third type of necessary knowledge includes, significantly enough, the authoritative accounts of events or facts which are geographically or historically remote, such a the existence of other countries, of historical personages, and of ancient kingdoms. To this type of knowledge belongs a supernatural or extraordinary variety, which God infuses directly into the Soul, without the help of intermediaries or sense organs, which are the normal channels of this type of knowledge.

The distinction between rational and authoritative knowledge was first broached by the Mu‘tazilah, who sought to extend the domain of reason well into regions which so far had been considered the exclusive preserve of revelation or faith. The Ash‘arite doctors, as illustrated in al-Baqalani's case, recognized the validity of rational knowledge but reacted instinctively against the Mu‘tazilite infringement on the domain of faith. On two fundamental questions of "natural theology" and ethics, namely, whether God can be known rationally, independently of revelation, and whether the knowledge of good and evil is possible prior to revelation, the Ash‘arite theologians took a qualified anti-Mu‘tazilite stand. The existence of God and His unity can be known rationally from the consideration of the createdness (Huduth) of the world and the logical necessity of a creator (muhdith).

To demonstrate this necessity, Ash‘arite doctors argued that the world, which they defined as everything other than God, was composed of atoms and accidents. Now accidents cannot endure for two successive moments, but are continually created by God, who produces and annihilates them at will. Similarly, the atoms in which these accidents inhere are continually created by God and can only endure by virtue of the accident of duration created in them by God. It follows from this premise that the world, being created, must necessarily have a creator.

Al-Baqalani's version of this argument differs little from the general Ash‘arite argument. He does, however, strengthen this argument by two others in which the "middle term" is different, but not the dialectical structure of the reasoning. In the first, he argues that the priority of certain things in time requires an "agent who made them prior," who is God. In the second, he introduces the concept of contingency and argues that things, considered in themselves, are susceptible of various forms or qualities. The fact that they actually possess certain forms and no others presupposes a "determinant" who decrees that they should receive these forms and no others, and this determinant is God. The last argument, or argument a contingentia mundi, is more fully developed by later authors, particularly al-Juwayni (d. 1086) in his al-Risalah al-Nizamiyah, and is the argument which, as we have seen, Ibn Sina fully exploited in his Metaphysics. It is noteworthy, however, that the generality of the Ash‘arite theologians showed a distinct predilection for the argument a novitate mundi (huduth) in so far as it harmonized with their concept of a world created in time by an omnipotent God.

On the other major issue of moral theology, the distinction between good and evil, the Ash‘arite doctors were equally in disagreement with the Mu‘tazilah. For, whereas the latter held that man can determine rationally what is good and evil, prior to revelation, the Ash‘arites adhered to a strict voluntarist ethics. Good is what God has prescribed, evil what He has prohibited. In keeping with this voluntarist thesis, they were reluctant to admit that any merit attached to that type of rational knowledge which is attainable through unaided reason. God's power and sovereignty are such that the very meaning of justice and injustice is bound up with His arbitary decrees. Apart from those decrees, justice and injustice, good and evil, have no meaning whatsoever. Thus God is not compelled, as the Mu‘tazilah had argued, to take note of what is "fitting" in regard to His creatures and to safeguard their moral or religious interests, so to speak, but is entirely free to punish the innocent and remit the sins of the wicked. And had He so desired, He could have created a universe entirely different from the one which He has in fact created, or refrained from creating this universe or any part of it altogether.

The metaphysical implementation of the theological and ethical outlook we have just outlined was the other major philosophical task the Ash‘arite school set itself. In this regard the differences between its major representatives, from al-Baqalani to al-Shahrastani, are minor. Al-Baqalani, however, played a pioneering role in elaborating the metaphysical groundwork of Ash‘arism. Significantly, later authors credit him with the introduction of atomism, which served as the metaphysical prop of Ash‘arite theology.

The introduction of atomism certainly antedates the rise of the Ash‘arite school itself, despite the statement of Ibn Khaldun that al-Baqalani was responsible for the "introduction of the rational premises on which proofs or theories depend, such as the existence of atoms, the void, and the proposition that an accident does not inhere in another accident or endure for two moments." From the accounts of Islamic atomism contained in the earliest treatise on Islamic "schisms and heresies," Maqalat al-Islamiyin, written by the founder of the Ash‘arite school himself, it appears that atomism had become firmly established in theological circles by the middle of the ninth century. Thus Dirar b. ‘Amr, a contemporary of wasil b. ‘Ata' (d. 748) and one of the earliest Mu'tazilite doctors of Basra, seems to have been the first theologian to challenge the generally accepted dualism of substance and accident. Al-Ash‘ari reports that Dirar held that "body is an aggregate of accidents, which once constituted, becomes the bearer of accidents." Similarly a thoroughgoing Shi'ite materialist who professed an anthropomorphic view of God of the crudest type, Hisham b. al-Hakam, challenged, as we have seen, this orthodox dualism and reduced everything to the notion of body, which according to him was divisible ad infinitum and consequently was not made up of atoms.

By the ninth century, the atomic theory of Kalam began to take definite shape. From al-Ash‘ari's account, we can infer that Abu'l-Hudhail (d. 841 or 849), al-Iskafi (d. 854-855), al-Juba'i (d. 915), al-Ash‘ari's own master, Mu‘ammar, a contemporary of Abu'l-Hudhail, as well as two contemporaries of his, Hisham al-Fuwati and 'Abbad b. Sulayman, accepted the atomic theory in one form or another. To take al-Juba'i as an instance, this doctor defined substance or the atom as the bearer or substratum of accidents, which, he added, "was such in itself, and can be conceived as substance prior to its coming-to-be," presumably in some disembodied Platonic state.

The metaphysical speculation on substance and accident, initiated by the Mu‘tazilah in the eighth century, was continued and refined by post-Mu‘tazilite doctors. The Ash‘arites, engrossed as they were with God's omnipotence and sovereignty in the world, found in atomism a convenient device for bolstering their theological claims. An Aristotelian world-view, dominated by causal processes that unfolded themselves almost mechanically, was ill-suited to their declared purpose of affirming God's prerogative to act freely and imperiously in the world. A collocation of atoms which depended, like the accidents inhering in them, on God's good pleasure, both for their creation and their duration, was more compatible in their view with the notion of God's arbitrary power .

Against the negators of the accidents, these doctors urged that the motion of a body subsequent to its rest is either due to the body itself or to something other than the body. The first alternative is absurd, since the body remains the same throughout the two successive states of motion and rest. Consequently it can only be due to something other than the body, which we call the accident. Similarly, the existence of a number of strokes inflicted by an agent on a patient, for instance, is distinct from the agent, the patient, or the instrument of striking. Therefore, the number of strokes is something distinct from all those factors, and that is what we understand by accident.

The number of the accidents which the orthodox recognized totals thirty. In a general way, they may be divided into primary and secondary accidents, depending on whether they accompany substance necessarily or not. The first of the primary accidents are the essential modi or states (singular: kaun) such as motion, rest, composition, location. Then come the accidents of color, heat, cold, etc. Al-Ash‘ari is reported by al-Baghdadi as holding that eight of the accidents accompany substance necessarily: motion, color, taste, smell, heat or its opposite, dampness or its opposite, life or its opposite, and finally duration.

The most peculiar variations on the theme of accidents are ascribed to Mu‘tazilite and Ash‘aite doctors. Thus the Mu‘tazilite al-Ka‘bi and his followers are said to have held thlat substance can be divested of all these "primary accidents" save color; and Abu Hashim, al-Juba'i's son, held that upon its coming into being, an atom can be divested of all accidents save the accident of being (kaun). Another Mu‘tazilite, al-Salihi, went a step further and argued that an atom could exist without any accidents whatsoever .

It is characteristic of these accidents, as al-Baghdadi relates, that they are not susceptible by themselves of any composition, contact, or transmission, since these are characteristics of the body alone. In this regard they were obviously analogous to the atoms, which were said by some theologians to be incapable by themselves of any composition, contact, or motion. However, the two are distinguished somehow, but theoretical difficulties persisted. Thus the Ash‘arite and, to some extent, the Mu‘tazilite doctors found the phenomena of motion quite baffling, and they resorted to the most far-fetched devices in attempting to explain motion rationally. Al-Nazzam, for instance, reduced every accident or quality, including human actions, to the universal category of motion, and even explained rest as a "motion of intention." Therefore, he argued, when a body is said to be static at a certain point, this can only mean that it had "moved in it twice." To account for the possibility of covering a certain distance, which consisted to him of an infinite number of points or particles, al-Nazzam introduced the concept of the leap (tafrah), or the view that a body could move from point A to point C without passing by the intermediary point B

The Ash‘arites, who subscribed to an even more extreme concept of discontinuous or discrete being, solved the difficulty in another way. They argued that motion and rest are two primary states or modi of substance, as has been noted. A substance which moves from one point to the other is at rest in relation to the second point, but in motion in relation to the first. Only al-Qalanisi, a somewhat dissident Ash‘arite, is reported by al-Baghdadi as holding that rest consisted of two successive states of being in the same place, whereas motion consisted of two successive states in the first and the second places necessarily

The most characteristic feature of the atoms of Kalam, as we have seen, was their perishable nature, which the Ash‘arites adhered to almost without exception. Not only al-Baqillani but the founder of the Ash‘arite school himself believed the accidents to be perishable by nature and to belong to the class of "transient things" (a‘rad) of this world, referred to in the Qur'an (8:67 and 46:24)

In demonstrating the perishability of accidents, al-Baghdadi argues that the "thesis of the durability of accidents entails their indestructibility. For if an accident is said to endure by itself…then it could persist in being until an opposite, necessitating its destruction, should come into being. However, there is no sufficient reason why such an opposite should arise and thereby counter its tendency to resist such an incursion."

Thus the duration of substances was made contingent upon the inherence in them of the accident of duration (baqa'). Since, however, this accident is not capable of duration per se, it followed that either the durability of substance is to be referred to other accidents of duration indefinitely, or else another principle of durability had to be introduced. This principle the Ash‘arites identified with God's own decree to preserve in being or destroy at will the atoms or ultimate components of physical objects in the world. Both the accidents and the atoms in which they inhere depended for their duration in this way on God's decree to repeat the process of their recreation as long as He pleased. Notwithstanding this circumstance, some Ash‘arite doctors found it necessary to give a rational account of a body's eventual corruption or annihilation. Thus al-Baqillani described annihilation (fana') as the act of withholding the two accidents of color and mode (kaun) from the body. Inasmuch as a body can never be divested of these two accidents, such an action necessarily entailed, according to him, the annihilation of the body. Such annihilation did not depend therefore on the inherence of the accident of corruption in the body, a thesis which, despite its strangeness, had at least one exponent. Al-Qalanisi argued that when God wishes to destroy a certain body, He creates in it the accident of corruption, which results in its destruction forthwith.

The contribution of late Ash‘arite doctors, such as al-Juwayni and al-Shahrastani, consists chiefly in elaborating or defending the concepts and methods to which the school as a whole was committed. The former, known also as Imam al-Haramayn, developed some of the epistemological and theological implications of Ash‘arite doctrine in al-Shamil, of which an abridgement, al-Irshad, was made by the author. Al-Shahrastani, an author of encyclopedic learning, wrote one of the best known and most comprehensive "heresiographies" in Arabic, K. al-Milal wa'l-Nihal. The second part of it is an invaluable source for the reconstruction of the Islamic picture of Greek philosophy. In addition, al-Shahrastani wrote a compendium of theology, Nihayat al-Iqdam, which surpasses many of the earlier treatises in its thoroughness and logical coherence, although it adds little to our knowledge of the scholastic tradition in theology .

III The Systematic Refutation of Neo-Platonism: Al-Ghazali

The greatest figure in the history of the Islamic reaction to Neo-Platonism is al-Ghazali, jurist, theologian, philosopher, and mystic. Born in Tus (Khurasan) in 1058, al-Ghazali addressed himself at an early age to the study of jurisprudence (fiqh) with a certain Radhkani, then moved on toJurjan, where he continued his studies with Abu'l-Qasim al-Isma'ili. His greatest teacher, however, was al-Juwayni, the outstanding Ash‘arite theologian of the period. Al-Juwayni initiated his brilliant pupil into the study of Kalam, philosophy, and logic. His introduction to the theory and practice of mysticism was due to al-Farmadhi (d. 1084), a renowned Sufi of the period.

Al-Ghazali's fortunes took a decisive turn as a result of meeting Nizam al-Mulk, vizier of the Saljuk sultan Malikshah. This able but doctrinaire vizier was fired by an intense zeal for the defense of Sunnite orthodoxy, and he consequently attacked the Shi‘ite (Isma‘ili) heterodoxy of the rival Fatimid caliphate at Cairo. The latter had so successfully wielded the double weapon of propaganda and political assassination throughout the Muslim world that the Saljuks felt compelled to reply in kind. To this end, Nizam al-Mulk set up a series of theological schools or seminaries, named after him, throughout the eastern part of the empire, where the study of Shafi'i fiqh and Ash‘arite theology were actively pursued. Al-Juwayni had been the head of the Nizamiyah of Nishapur until his death in 1085. It now devolved upon his disciple to Serve the cause of Sunnite orthodoxy.

For five years (1091 to 1095), then, al-Ghazali, as head of the Nizamiyah of Baghdad, pursued his teaching in jurisprudence and theology with great success. The troubled political situation of the times and the violent death of Nizam al-Mulk in 1092 at the hand of an Isma'ili assassin, followed shortly after by the death of the sultan Malikshah, appear to have contributed to his gradual disillusionment with teaching. His initiation into the practice of the Sufi way, between 1093 and 1094, no doubt added to his sense of the futility of a career that was not dedicated to the disinterested quest of truth or the service of God.

In a moving autobiographical work, al-Munqidh, which has been compared to St. Augustine's Confessions, al-Ghazali tells the dramatic story of his spiritual and intellectual anxiety and doubt; his renunciation, at the height of his fame, of his' teaching career at Baghdad in 1095; his peregrinations throughout Syria, Palestine, and Hijaz; and his eventual resumption of teaching, eleven years later, at Nishapur. This second term of instruction, however, was short lived. Five years later, in 1111, his eventful and active life as a scholar and mystic came to an untimely end.

Al-Ghazali's autobiography introduces us, almost from the very first line, to the intellectual and spiritual problems with which he had to contend throughout his whole life, and particularly during the period of tribulation which followed his resignation from the Nizamiyah school at Baghdad. Even before he was twenty, al-Ghazali tells us, he had been seized by an ardent desire for truth and had been distressed at the spectacle or conflicting beliefs and creeds and the passivity and credulity of the common run of mankind who defer blindly to the authority of their elders. Accordingly, he resolved to search for "certain knowledge," which he defines as "that knowledge in which the object is known in a manner which is not open to doubt at all," so that if its truth were to be challenged by a miracle-maker, it would withstand that challenge. When he proceeded to inquire whether he was actually in possession of such knowledge, he was led to conclude that the only knowledge which tallied with this description was sense knowledge and the knowledge of self-evident propositions. In order to pursue the process of doubt to its logical consequence, however, he felt he had to satisfy himself that such knowledge was indeed certain. At the end of a painful process of doubt, he found that in fact it was not. For, in the case of the former, our senses often judge that the object is such and such, but their judgment is soon subverted by reason. For instance, we look at a shadow and infer that it is stationary, but soon after we are compelled to admit that it was not. Or we look at a remote object, such as the planet, which appears to our senses to be the size of a coin, whereas astronomical evidence compels us to believe that it is many times larger than the earth.

If sense experience is not to be trusted, then by analogy the knowledge of necessary propositions or axioms is not to be trusted either. For, as the senses at once reminded al-Ghazali: What right had he to think that his confidence in the necessary propositions of reason differed from his confidence in sensible knowledge? The latter had been shown by reason to be doubtful; might it not then be that there "exists beyond reason a higher authority, which would, upon its manifestation, show the judgment of reason to be invalid, just as the authority of reason had shown the judgment of sense to be invalid?" The analogy of dream is instructive here. Very often in dreams we are confident of the reality of our experiences but this confidence is dispelled as soon as we wake up. Might it not be, then, that our waking life is no better, as the Prophet has said, than dreaming, in comparison with the life after death?

These doubts, al-Ghazali tells us, continued to afflict him like a real sickness for almost two months. Eventually he recovered his intellectual health, not through his own efforts, but rather through a "light which God infused into his heart, which indeed is the key to most species of knowledge." This light, he now realized, was not a matter of discourse or argument, but of divine grace, which the Prophet had described as "the dilation of the heart, whereby it becomes prone to the reception of Islam." The signs attendant upon it are the renunciation of this world of illusion and the turning toward the world of reality.

Much has been written about al-Ghazali's sincerity and the significance of his use of the method of doubt. Whether or not the account given in al-Munqidh is a factual record of his spiritual and intellectual experience is a purely academic question. What is of particular significance is the profound earnestness with which he depicts in this work the "states of his own soul" as it was assailed by doubt, recovered faith through the outpouring of divine light, and how finally he consented to champion publicly the cause of orthodoxy against the sectarians of heresy and deceit.

Of these sectarians he singles out four groups that might be presumed to be in possession of the (Islamic) truth in the eleventh century; if none of them was in possession of such truth, the quest for certainty would be entirely futile. These four are the theologians, the Isma'ilis (or Batinis), the philosophers, and the Sufis.

The aim of theology (Kalam), which he had first studied, was the defense of orthodoxy and the repulsion of the heretics' attacks on it. In this defense, the theologians start with some premises that are not certain in themselves but must be accepted on the authority of Scripture or the Consensus of the community. Hence, this branch of learning, though useful, does not lead per se to that indubitable certainty which al-Ghazali was seeking.

The Isma'ili doctrine, known as Ta‘lim (instruction) during this period, did not quench his thirst for truth either; For the substance of Isma'ili doctrine is that the knowledge of truth is not possible without a teacher, and the only teacher whose teaching cannot be doubted is an infallible teacher, or, as he was called by the Isma'ilis, the Imam. Here, however, the question arises: What are the marks of such an infallible Imam and where is he to be found? Muslims have an infallible teacher, namely, the Prophet. The Isma'ilis Who recognize the authority of the Prophet argue nevertheless that he is dead. In reply to this argument it can be urged that the Imam, though not dead, is equally inaccessible, since he is said to be m temporary concealment(gha'ib).

Despite the vehemence with which al-Ghazali inveighs against the Isma'ilis and their splinter groups in his works, his polemics against the Arab Neo-Platonists are by far the most sustained and the most searching. And it is naturally these polemics that are particularly interesting to us here. The suppressed, almost instinctive, reaction against rationalism in general and Greek philosophy in particular, which had been a characteristic mark of orthodoxy heretofore, bursts forth in al-Ghazali's attacks on the Muslim Neo-Platonists, particularly al-Farabi and Ibn Sina. Earlier orthodox writers had been content to challenge rationalism or to reproach the philosophically inclined, on grounds either of piety or of xenophobia. Al-Ghazali, who agreed with the general sentiment of the orthodox, felt nonetheless that "only one who has mastered the science [of philosophy] to such a degree that he can vie with the most proficient in that science" and even excel them will be qualified to show the incoherence of their doctrine. Since no one had accomplished this difficult task before him, al-Ghazali felt compelled to grapple with this problem with all his might. He therefore turned to the study of philosophy in his spare time, since he was occupied during this period with teaching religious subjects to no fewer than 300 students at the Nizamiyah of Baghdad. Although he does not mention this in his autobiography, he had, as we have seen, already made a start in that direction as a student of al-Juwayni in Nishapur. Presumably, his study of philosophy in a systematic way was made during this second period. In three years, he was able, according to his own account, "through God's assistance," to master the philosophical sciences completely. The fruit of these years of philosophical initiation was a work entitled the Intentions of the Philosophers, in which he states that his express purpose is to expound the doctrines of the philosophers, as a prelude to refuting them in a subsequent work. This exposition of the tenets of Arab Neo-Platonism is so skillfully written that a careless reader would conclude that it is the work of a conventional Neo-Platonist, as indeed the thirteenth-century Scholastic doctors had concluded when it appeared in the Latin version of Dominicus Gundissalinus, entitled Logica et Philosophia Algazelis Arabis. The circumstance was at the root of the widspread belief in the later Middle Ages that Algazel was a genuine Neo-Platoinist of the stamp of Avicenna and others.

Other fruits of al-Gazali's philosophical intition are to be found in an important manual of Aristotelian logic, Mi'yar al-'Ilm (The criterion of Science). This work, and the Intentions and Tahafut, form a philosophical trilogy of the utmost significance for the study of history of the struggle between the theologians and the philosophers of Islam.

We are not, however, concerned here with al-Ghazali's contribution to the dissemination of Neo-Platonism, since his professed aim was not its advancement but rather its rebuttal. Indeed, al-Farabi and Ibn Sina, the two major targets of his attack, had by thoroughness rendered any further creative contribution in that domain almost impossible. The chief contribution of al-Ghazali lay instead in his identification with the antiphilosophical party, and his attempt to prove the incoherence of the philosophers on philosophical grounds; hence his importance in the history of philosophical thought in Islam.

Al-Ghazali's motive in writing his Tahafut (or Collapse of the Philosophers) is stated explicitly to be religious. What prompted him write this work, he tells us, was the way in which as small group of free-thinkers had been led to repudiate Islamic Beliefs and neglect the ritual basis of worship as unworthy of their intellectual attainments. They were confirmed in this by the widespread adulation reserved for ancient philosophers, from Socrates to Aristotle, who were erroneously supposed to partake of their irreligion. However, had they taken the trouble to examine the teaching of those philosophers, they would have discovered that "every one worthy of note among the ancients and the moderns" subscribed to the fundamentals of religious belief, i.e. the existence of God and the reality of the Day of judgement. Differences among them affect only incidentally the substance of their belief.

In substantiating the latter claim, al-Ghazali draws a distinction between those philosophical sciences such as mathematics and logic, which are completely innocuous from a religious point of view, and those which, like physics and metaphysics, contain the bulk of the heresies or errors of the philosophers. Three of those philosophers deserve special mention: Aristotle, who organized and perfected the philosophical sciences; and al-Farabi and Ibn Sina, who are the two most authoritative and trustworthy expositors of Aristotelian philosophy in Islam. The rebuttal of the view of these three should enable the critic of philosophy to dispense with the rebuttal of lesser figures.

Al-Ghazali's attack is thus judiciously leveled at the two leading Muslim Neo-Platonists directly, and indirectly at Aristotle, their master. Altogether, he enumerates sixteen metaphysical and four physical propositions that have an obvious religious relevance and against which the unguarded believer must be warned. Of these propositions, three are particularly obnoxious from a religious point of view, and consequently those who uphold them must be declared renegades, liable to the religious sanctions against renegades in Islam. These propositions are the eternity of the world a parte ante, God's knowledge of universals only, and the denial of the resurrection of the body. The remaining seventeen propositions do not, in al-Ghazali's opinion, justify the charge of irreligion (kufr), but simply that of heresy (bid‘ah). Many of them are professed by other sectarians of Muslim heresay, such as the Mu‘tazilah, and should not on that account be regarded as equivalent to apostasy except on a very narrow-minded or bigoted interpretation, which al-Ghazali is careful to disallow.

The first proposition of the Tahafut bears on the eternity of the world as professed by the Islamic Neo-Platonists and Aristotle. In their espousal of the emanationist world-view, as we have seen, the latter had disassociated themselves from the main body of orthodox Islam. As early as al-Ash‘ari, the heterodox implications of the thesis of eternity had been clearly discerned by the theologians, but with the exception of Ibn Hazm (d. 1064) no systematic exposition and refutation of these implications had been attempted before al-Ghazali's time. Implicit in the polemic of the theologians against this thesis is the claim that it militated against the Qur'anic concept of creation ex nihilo, and as a corollary involved an arbitrary limitation of God's absolute power.

The views of the philosophers on the question of the eternity of the world are stated by al-Ghazali to be three: (1) the view of the vast majority, ancient and modern, who believed it to be eternal; (2) the view of Plato, who held that it was created in time; and (3) the view of Galen, who suspend judgment on this issue.

In his rebuttal of the eternalist thesis, al-Ghazali asserts that the world was created in time, through an eternal decree of God. He rejects in this connection the claim that the lapse of time which separates the eternal decree of God and the creation of the world involves the supposition that God could not accomplish the creation at once. This claim, he argues, does not rest on any demonstrative grounds but is simply a dogmatic assertion.

A mathematical argument is then advance against the Neo-Platonists. The eternity of the world entails logically that an infinite number of revolutions of the heavens have already elapsed. We know, however, that these revolutions can serve as the basis of mathematical computations. For instance, the sphere of the sun completes a single revolution in a year, that of Saturn in 30, that of Jupiter in 12, and that of the firmament in 36,000 years. A finite ratio between the revolutions of the sun and the other spheres can be given as follows: 1/30, 1/12, 1/36,000 respectively, which would contradict the assumption that these revolutions are infinite and occur in an infinite time.

Moreover, these revolutions are either odd or even, and must consequently be finite. For the infinite is neither odd nor even, since it can be increased by one indefinitely, while remaining infinite. To top it all, the Neo-Platonists assert the possibility of an infinite number of Souls, existing in a disembodied condition, as Ibn Sina held, despite the logical contradiction which the concept of an actual infinite involves.

In his rebuttal of the Avicennian arguments that God is prior to the world in essence, rather than in time, al-Ghazali takes an unequivocal stand in support of the creation of time. When we say that God is prior to the world, we simply mean, according to him, that God existed while the world was not, and continued to exist together with the world. What these two propositions assert is the existence of an entity (God) followed by both entities together. The representation of a tertium quid (time) is a trick of the imagination, which compels us to represent both entities as linked together, through this tertium quid.

As for the view that prior to its creation the world was obviously possible, it does not necessarily entail, as the Neo-Platonists contend, an eternal substratum in which possibility inheres. For, on this view, not only the possible but its two contraries, the impossible and the necessary, would also require such a substratum, and this is clearly absurd. The possible, the impossible, and the necessary, as indeed all other common qualities, have only a conceptual reality. What exists is simply the entity of which they are predicated.

The second question of the Tahafut deals with perpetuity or eternity, of which post-eternity is explicitly stated to be a logical this question raise the same crucial theological issues as that of pre-eternity, of which post-eternity is explicitly stated to be a logical offshoot. A whole group of questions (3 to 11) deals next with God and His attributes. In Question 3, the fundamental issue is raised whether, in the context of Neo-Platonism, God could be rightly described as the Creator or Maker of the world. For, according to the Neo-Platonists, the world emanates from God (or the First, as they call Him) necessarily, just as the effect emanates from the cause or the light from the sun. Now a genuine agent must be conscious and free, so that God can only be designated by these philosophers as the Maker (Sani‘) of the world metaphorically.

Moreover the world, being eternal, according to them, can hardly be said to be created. For creation or making denotes the act of bringing an entity forth into being, out of nothing, and the eternal is forever in being. Likewise the Neo-Platonists hold that out of one only one can come (ex uno non fit nisi unum), but since God is one and the world multiple, there can be no sense in saying that He is its Maker. Indeed, from their premises it would follow that only a series of ones or simple entities could emanate from the "First." As to the multiple or composite entities which make up the world, none of the arguments of the Neo-Platonists can account for their production.

What is more, the Neo-Platonists are unable to prove the existence of God either (Question 4). All their arguments rest on the impossibility of an infinite egress and the necessity of positing ultimately an Uncaused Cause of the series of effects. However, (a) bodies are eternal, according to them, and require, in consequence, no cause, and (b) an infinite series is not impossible since it follows from their thesis of the eternity of the world that an infinite series of effects has come and gone heretofore. Some of them, as we have seen in the case of Ibn Sina, even admit that an infinite number of Souls can exist in a disembodied condition.

Al-Ghazali next turns to the question of divine attributes. The Neo-Platonists are unable prove the unity of God (Question 5). The substance of their proof is that if we posit two necessary beings, necessity would not belong to each of them essentially, but through a cause, so that the Necessary Being would be caused, which is absurd. This proof is not valid because their distinction between the necessary-in-itself and the necessary-through-a-cause, upon which this proof rest, is unfounded. The Neo-Platonists, in fact, deny, the divine attributes altogether (Question 6). Such attributes are, according to them, accidents of the essence and, as such, involve plurality and contingency in the subject, God, they claim, cannot be the bearer often attributes, but they admit at the same time that He is nonetheless knowing (‘alim), which obviously implies that He possesses the attribute of knowledge, however we might interpret it.

The question of divine knowledge is the second issue on which Al-Ghazali denounces the Neo-Platonists. We might pause therefore to consider his objections at length. In Question 11 he introduces the discussion by expounding the Islamic (Ash‘arite) view of divine knowledge, Since the act of willing implies the knowledge of what is willed, and the whole world has been willed by God, it follows that the whole world is known to Him and is caused by this double act of knowing and willing. But to be capable of knowledge and will is to be alive. Therefore God must be alive and, as such, capable of knowing everything which emanates from Him, together with Himself, as its source. The Neo-Platonists, who have stripped God of all essential attributes, have been led to conclude that "the Lord of Lords and the Cause of Causes has no knowledge whatsoever of anything which happens in the world. [One might ask them, therefore] what difference is there between Him and the dead, except in regard of His self-knowledge [ which they admit], and what excellence does this self-knowledge involve, when coupled with ignorance of everything else?"

The philosophers, having thus denied that God has will, are unable to prove that He has knowledge either. The substance of the argument of Ibn Sina on this score, for instance, is that the First, being entirely immaterial, must be a pure intellect (‘aql), and must accordingly know all things, since the only bar to such knowledge is matter. However, Ibn Sina and his fellow Neo-Platonists are unable to substantiate their claim that God is an intellect, but simply infer it from the premise that He is not a material entity. However, all that can be inferred from the proposition that the First is not a material object, nor an accident of a material object, is that He is self-subsistent. To argue that He is in consequence an intellect, since He knows Himself or knows other things, is to beg the question. Only on the assumption that He knows Himself, as well as other things, can it be asserted that He is a pure intellect, which is precisely the point at issue.

It may be objected that the philosophers do not deny that the world is the product of God's action, but only that He has willed it in time. For they do not question the general proposition that the agent is necessarily conscious of his action. God, who has produced the "whole," must in consequence be conscious of His production.

Al-Ghazali counters this objection on three grounds. (1) The world is said by the Neo-Platonists to emanate from God by a "necessity of nature," analogous to the emanation of light from the sun. Obviously, such an emanation does not involve either willing or thinking on the part of the agent. (2) Some (e.g., Ibn Sina) claim that the emanation of the "whole" from God is the result of His knowledge of this "whole" and this knowledge is identical with the essence of God. This claim, however, is disputed by other philosophers, who describe emanation in terms of natural necessity, as in (1) .(3) Even if the latter version of emanation is accepted, the only corollary thereof is that God knows only the first entity to result from His action, i.e., the first intellect, which in turn knows what results from it, and so on down the scale of subsequent emanations. God cannot, according to this version, know the "whole" either.

Indeed, it does not follow from the premises of the philosophers that God knows Himself either (Question 12), for we infer the knowledge of self from the fact of life, which in turn is inferred from knowledge and will. The philosophers, in denying that God is capable of willing, as we have seen, are unable to prove that He knows Himself or anything that follows from Him. To be consistent, the philosophers must deny that God is capable of knowing, seeing, or hearing (attributed to Him by the generality of Muslims), since these attributes denote, according to them, imperfections rightly predicated of the creature but not of God.

Perhaps the most crucial aspect of the problem of divine knowledge, from the Islamic point of view, is the denial of God's knowledge of particulars. The Qur'an states explicitly (e.g., 34:3) that nothing escapes God's knowledge, not even "the smallest particle in heaven or on earth." Philosophers who admit, like Ibn Sina, that God knows things other than Himself, have argued nevertheless that the mode of His knowledge is "universal." It is not subject, like "particular" knowledge, to the limitations of time or place. Thus God knows an event (say, the eclipse of the sun) prior to its occurrence or subsequently, in the same instantaneous manner. For He knows a priori the series of causes from which it will ultimately result. Similarly, He knows an individual man, for instance Zaid or Amr, in so far as He knows the "absolute man," i.e., independently of the conditions of time or place. The particular or accidental qualities, or the spatiotemporal determinations, which set such an individual apart from other individuals, are objects of sense experience of which God cannot possibly partake.

In his rebuttal al-Ghazali argues that God's knowledge is indeed independent of the conditions of time and space. It does not, on that account, exclude relation to particulars, which are subject to such conditions. The changes to which the mode of this knowledge is liable do not involve change in the essence of the knower, but rather in the relationship of his knowledge to the object, which is continually changing.

If it is maintained nonetheless that such relations enter into the definition of the object, so that change in the latter will involve change in the knower necessarily, one might retort that, if this were true, even the knowledge of universals would involve change in the knower, in so far as such knowledge involves different relationships to the knower. And since these universals are infinite in number, it is not clear how, on the argument of the philosophers, the unity of God's knowledge can be safeguarded, unless we assume that the change or plurality in the object does not necessarily affect the knower.

Nor does it follow from the premises of the Neo-Platonists that the Eternal Being (God) is not subject to change. They posit that the world (which according to them is eternal) is nevertheless subject to change. As to the detraction from the perfection of God which the dependence of His knowledge upon the changing object must involve, we can only observe that there can be no greater detraction from this perfection than the claim of the philosophers that everything emanates from God by way of natural necessity, without His knowledge or preordination.

In the "physical" part of the Tahafut, al-Ghazali considers two major questions: the repudiation of the necessity of the causal nexus and the resurrection of the body. The former (Question 17) had been one of the major issues which more than two centuries earlier had set the theologians against the philosophers in general and the Peripatetics in particular. The tendency of the latter to ascribe to "secondary causes" a certain degree of efficacy in the natural order was frowned upon by the theologians on the ground that it militated against the Qur'anic concept of an omnipotent Deity who carried out His grand cosmic designs imperiously and directly and who, in consequence, had no need of any mediator. The occasionalist metaphysics of atoms and accidents, which as we have seen was developed by the theologians of the ninth century, was designed precisely to safeguard God's absolute independence from any conditions or limitations, natural or other. With the exception of a few Mu‘tazilite theologians who introduced the concept of generation (tawallud) as a theoretical device for retaining the efficacy of natural agents, the Muslim theologians rejected "secondary causation" as incompatible with God's uniqueness and sovereignty in the world. Al-Ghazali, however, was the first theologian to undertake a systematic refutation of the concept of a necessary causal nexus. In this, he appears to have been influenced by the Greek skeptics of the Pyrrhonian school.

The discussion of causality opens with the statement that the correlation between the so-called cause and effect is not necessary, for only where logical implication is involved can a necessary correlation be admitted. It is plain, however, that between two distinct conditions or events, such as eating and satiety; contact with fire and burning, decapitation and dying, no such correlation can be asserted. The observed correlation between concomitant events in medicine, astronomy, and the arts is due merely to God's action in joining them constantly. It is logically possible, however, for this conjunction to be infringed and the so-called effects be produced ab initio, without their concomitant causes, as indeed happens in what Muslims universally regard as miracles.

Take the case of fire in relation to cotton. The philosophers claim that fire causes the burning of the cotton, whereas we maintain, says al-Ghazali, that the real agent in this process is God, acting either directly by Himself, or indirectly through an angel. For fire is inanimate, and cannot therefore be said to cause anything whatsoever. The only proof that the philosophers can advance is that we observe burning to occur upon contact with fire, but observation simply proves that the burning follows upon contact with fire, not that it is due to it, or that it is in fact the only possible cause of burning.

Or take the case of life and growth, in relation to the animal. It is plain that life, as well as the cognitive and motive faculties which inhere in the sperm of the animal, are not the effects of the four primary qualities. Nor is the father, who deposits the sperm in the mother's womb, the cause of the infant's life, hearing, seeing, etc. This cause is the First Being. In fact, the major philosophers admit that the accidents or events which result from the conjunction of natural causes and effects are ultimately due to the "Giver of Forms," who is an angel or a separate substance from whom the "substantial forms" of natural objects emanate, once matter has become sufficiently disposed to receive them.

However, the philosophers might admit that the ultimate causes of natural processes are supermundane, and yet ascribe to the action of natural causes or agents the disposition or aptitude for receiving their action. Accordingly, if we posit that fire is of a certain nature and cotton is of a certain nature also, it is impossible that fire should sometimes burn cotton and sometimes not, unless the nature of fire or that of cotton has changed in the interval.

Al-Ghazali's solution of this difficulty is that the supermundane principles or agents, particularly God, do not act by way of causal necessity, as the philosophers claim, but rather by way of will. Consequently, it is quite possible logically for God to cause burning in some instances but not in others. One might object that, on this supposition, everything becomes possible and nothing will be known with certainty, except where God wished at the same time to impart directly the knowledge corresponding to the action. For instance, we may imagine a man looking out on a strange scene: fire burning, lions roaring, soldiers marching, without beholding any part of it, because God did not create in him the corresponding perception of this scene at the time. Or we may leave a book behind and, on returning home, find that the book has changed into a lad or a beast, or the lad has turned into a dog, etc. God could thus create whatever He pleases, in any order He pleases, since He is not bound by any order, causal or other .

In his retort, al-Ghazali states that these absurdities would result only if we assume that God will not create in us the knowledge corresponding to the events or to the fact that they are possible. But God has created in us the knowledge that these events are merely possible, not that they are actual. They could just as well occur as not occur. Their repetitiveness "establishes firmly, in our minds [the notion] of their occurrence according to the past habitual course." But it is possible for a prophet or an ordinary man with prophetic or acute intuitive powers to foresee that such events will happen in a manner which does not conform with the normal course of events in nature. In such situations God simply creates in the mind of the knower the corresponding knowledge and thereby the alleged difficulty vanishes.

The knowledge of the sequence of such events is normally dependent on their actual occurrence. Without denying that certain elements, e.g., fire, are endowed with certain properties, such as the power to burn cotton, however, it is not logically excluded that God or His angels may cause this power to be checked in such a way that it will not cause burning in the cotton; or He may create in the cotton the power to resist the action of burning. Such miracles, reported in the Qur'an, as [Christ's] resurrecting the dead or [Moses'] turning a stick into a serpent could thus be explained in a perfectly rational manner. Or it may be possible for God to effect His miraculous designs without violence to the natural process of events, but rather through what might be called telescoping or abridging this process. Thus matter, according to the Peripatetic philosophers, is susceptible of many contrary qualities. The generation of animals, in their view, results from a series of permutations culminating in the animal in question. Earth turns into a vegetable, which upon being consumed by the animal parent turns into blood, which in turn is converted inside the body into the seminal fluid, which eventually develops into an individual offspring. Habitually, this process takes a fairly long time, but it is not logically excluded that God could bring these permutations about in a shorter period than is His wont, and then in progressively shorter periods until we come to a period so short as to be instantaneous. And this is what we denote as a miracle.

Indeed the philosophers allow that the generation of animals or vegetables is bound up with the ability of matter, as it becomes disposed through the influence of the heavenly conjunctions and their diverse motions, to receive the forms which emanate from the active intellect or Giver of Forms, as we have seen. On this supposition, the most extraordinary occurrences in. the world become possible, and extraordinary events or miracles perfectly intelligible.

The final three "physical" questions of the Tahfut deal with the nature of the Soul and its immortality, according to Neo-Platonic doctrine. In Question 18, al-Ghazali sets forth the arguments of those philosophers for the immateriality and simplicity of the Soul and shows that they are simply inconclusive. Nor are their arguments for its immortality conclusive either, for these rest on the simplicity and the immateriality of the Soul, which they are unable to establish (Question 19).

Since none of these arguments is conclusive, the only recourse left is the authority of Scripture or revelation (al-shar‘), which asserts immortality in an undoubted manner and expatiates on the state of the Soul in the after-life. Much of what the philosophers say concerning the noncorporeal or spiritual pleasures reserved to the Soul in the after-life is in conformity with the teaching of Scripture. What we question, al-Ghazali argues, is that their knowledge of the immortality of the Soul and the spiritual pleasures or pains of the after-life are known through unaided reason, and that they are the only types of pleasure or pain which man can experience after death. There is no logical absurdity involved in positing both types of pleasure or pain, i.e., the spiritual and the bodily, as well as the bodily resurrection laid down in Scripture. The claim of the philosophers that sensuous pleasures and pains, as depicted in the Qur'an, are no more than allegories intended for the edification of the masses is very tenuous, and the analogy between the passages in the Qur'an that describe them and the passages that refer to God in anthropomorphic terms is not a sound analogy. The latter can and ought to be interpreted allegorically, but not the former. For, whereas it is logically impossible that God should be described in corporeal terms and as possessing physical members or occupying space, the bodily rewards and punishments alluded to in the Qur'an are not logically impossible, as has been shown under the general heading of miracle and the miraculous. God could thus restore the Soul on the Day of Judgment to a body either identical with or analogous to its original body, and thereby enable it to partake of both bodily and nonbodily pleasure. In fact, it is with the possibility of such a dual enjoyment that its complete happiness is bound up.