(The Incoherence of the Incoherence)

by
Ibn Rushd

Translated from the Arabic with Introduction and Notes
by
Simon Van Den Bergh

Volume I

Introduction

If it may be said that Santa Maria sopra Minerva is a symbol of our European culture, it should not be forgotten that the mosque also was built on the Greek temple. But whereas in Christian Western theology there was a gradual and indirect infiltration of Greek, and especially Aristotelian ideas, so that it may be said that finally Thomas Aquinas baptized Aristotle, the impact on Islam was sudden, violent, and short. The great conquests by the Arabs took place in the seventh century when the Arabs first came into contact with the Hellenistic world. At that time Hellenistic culture was still alive; Alexandria in Egypt, certain towns in Syria — Edessa for instance — were centres of Hellenistic learning, and in the cloisters of Syria and Mesopotamia not only Theology was studied but Science and Philosophy also were cultivated. In Philosophy Aristotle was still ‘the master of those who know’, and especially his logical works as interpreted by the Neoplatonic commentators were studied intensively. But also many Neoplatonic and Neopythagorean writings were still known, and also, very probably, some of the old Stoic concepts and problems were still alive and discussed.

The great period of translation of Greek into Arabic, mostly through the intermediary of Christian Syrians, was between the years 750 and 850, but already before that time there was an impact of Greek ideas on Muslim theology. The first speculative theologians in Islam are called Mu‘tazilites (from about A. D. 723), an exact translation of the Greek word σχισματικοί (the general name for speculative theologians is Mutakallimun, διαλεκτικοί, dialecticians, a name often given in later Greek philosophy to the Stoics). Although they form rather a heterogeneous group of thinkers whose theories are syncretistic, that is taken from different Greek sources with a preponderance of Stoic ideas, they have certain points in common, principally their theory, taken from the Stoics, of the rationality of religion (which is for them identical with Islam), of a lumen naturale which burns in the heart of every man, and the optimistic view of a rational God who has created the best of all possible worlds for the greatest good of man who occupies the central place in the universe. They touch upon certain difficult problems that were perceived by the Greeks. The paradoxes of Zeno concerning movement and the infinite divisibility of space and time hold their attention, and the subtle problem of the status of the nonexistent, a problem long neglected in modern philosophy, but revived by the school of Brentano, especially by Meinong, which caused an endless controversy amongst the Stoics, is also much debated by them.

A later generation of theologians, the Ash‘arites, named after Al Ash‘ari, born A. D. 873, are forced by the weight of evidence to admit a certain irrationality in theological concepts, and their philosophical speculations, largely based on Stoicism, are strongly mixed with Sceptical theories. They hold the middle way between the traditionalists who want to forbid all reasoning on religious matters and those who affirm that reason unaided by revelation is capable of attaining religious truths. Since Ghazali founds his attack against the philosophers on Ash‘arite principles, we may consider for a moment some of their theories. The difference between the Ash‘arite and Mu‘tazilite conceptions of God cannot be better expressed than by the following passage which is found twice in Ghazali (in his Golden Means of Dogmatics and his Vivification of Theology) and to which by tradition is ascribed the breach between Al Ash‘ari and the Mu‘tazilites.

‘Let us imagine a child and a grown-up in Heaven who both died in the True Faith, but the grown-up has a higher place than the child. And the child will ask God, “Why did you give that man a higher place?” And God will answer, “He has done many good works.” Then the child will say, “Why did you let me die so soon so that I was prevented from doing good?” God will answer, “I knew that you would grow up a sinner, therefore it was better that you should die a child.” Then a cry goes up from the damned in the depths of Hell, “Why, O Lord, did you not let us die before we became sinners?”’

Ghazali adds to this: ‘the imponderable decisions of God cannot be weighed by the scales of reason and Mu‘tazilism’.

According to the Ash‘arites, therefore, right and wrong are human concepts and cannot be applied to God. ‘Cui mali nihil est nec esse potest quid huic opus est dilectu bonorum et malorum?’ is the argument of the Sceptic Carneades expressed by Cicero (De natura deorum, iii. 15. 38). It is a dangerous theory for the theologians, because it severs the moral relationship between God and man and therefore it cannot be and is not consistently applied by the Ash‘arites and Ghazali.

The Ash‘arites have taken over from the Stoics their epistemology, their sensationalism, their nominalism, their materialism. Some details of this epistemology are given by Ghazali in his autobiography: the clearness of representations is the criterion for their truth; the soul at birth is a blank on which the sensations are imprinted; at the seventh year of a man’s life he acquires the rational knowledge of right and wrong. Stoic influence on Islamic theology is overwhelming. Of Stoic origin, for instance, are the division of the acts of man into five classes; the importance placed on the motive of an act when judging its moral character; the theory of the two categories of substance and accident (the two other categories, condition and relation, are not considered by the Muslim theologians to pertain to reality, since they are subjective); above all, the fatalism and determinism in Islam which is often regarded as a feature of the Oriental soul. In the Qur’an, however, there is no definite theory about free will. Muhammad was not a philosopher. The definition of will in man given by the Ash‘arites, as the instrument of unalterable fate and the unalterable law of God, is Stoic both in idea and expression. (I have discussed several other theories in my notes.)

Sometimes, however, the theologians prefer to the Stoic view the view of their adversaries. For instance, concerning the discussion between Neoplatonism and Stoicism whether there is a moral obligation resting on God and man relative to animals, Islam answers with the Neoplatonists in the affirmative (Spinoza, that Stoic Cartesian, will give, in his Ethica, the negative Stoic answer).

The culmination of the philosophy of Islam was in the tenth and eleventh centuries. This was the age also of the great theologians. It was with Greek ideas, taken in part from Stoics and Sceptics, that the theologians tried to refute the ideas of the philosophers. The philosophers themselves were followers of Aristotle as seen through the eyes of his Neoplatonic commentators. This Neoplatonic interpretation of Aristotle, although it gives a mystical character to his philosophy which is alien to it, has a certain justification in the fact that there are in his philosophy many elements of the theory of his master Plato, which lend themselves to a Neoplatonic conception. Plotinus regarded himself as nothing but the commentator of Plato and Aristotle, and in his school the identity of view of these two great masters was affirmed. In the struggle in Islam between Philosophy and Theology, Philosophy was defeated, and the final blow to the philosophers was given in Ghazali’s attack on Philosophy which in substance is incorporated in Averroës’ book and which he tries to refute.

Ghazali, who was born in the middle of the eleventh century, is one of the most remarkable and at the same time most enigmatic figures in Islam. Like St. Augustine, with whom he is often compared, he has told us in his autobiography how he had to pass through a period of despair and scepticism until God, not through demonstration but by the light of His grace, had given him peace and certitude. This divine light, says Ghazali, is the basis of most of our knowledge and, he adds, profoundly, one cannot find proofs for the premisses of knowledge; the premisses are there and one looks for the reasons, but they cannot be found. Certitude is reached, he says, not through scholastic reasoning, not through philosophy, but through mystical illumination and the mystical way of life. Still Ghazali is not only a mystic, he is a great dogmatist and moralist. He is regarded as Islam’s greatest theologian and, through some of his books, as a defender of Orthodoxy. It is generally believed that the Tahafut, the book in which he criticizes Philosophy, was written in the period of his doubts. The book, however, is a Defence of Faith, and though it is more negative than positive, for it aims to destroy and not to construct, it is based on the theories of his immediate predecessors, many of whose arguments he reproduces. Besides, he promises in this book to give in another book the correct dogmatic answers. The treatise to which he seems to refer does not contain anything but the old theological articles of faith and the Ash‘arite arguments and solutions. But we should not look for consistency in Ghazali; necessarily his mysticism comes into conflict with his dogmatism and he himself has been strongly influenced by the philosophers, especially by Avicenna, and in many works he comes very near to the Neoplatonic theories which he criticizes. On the whole it would seem to me that Ghazali in his attack on the philosophers has taken from the vast arsenal of Ash‘arite dialectical arguments those appropriate to the special point under discussion, regardless of whether they are destructive also of some of the views he holds.

Averroës was the last great philosopher in Islam in the twelfth century, and is the most scholarly and scrupulous commentator of Aristotle. He is far better known in Europe than in the Orient, where few of his works are still in existence and where he had no influence, he being the last great philosopher of his culture. Renan, who wrote a big book about him, Averroes et l’Averro’asme, had never seen a line of Arabic by him. Lately some of his works have been edited in Arabic, for instance his Tahafut al Tahafut, in a most exemplary manner. Averroës’ influence on European thought during the Middle Ages and the Renaissance has been immense.

The name of Ghazali’s book in which he attacks the philosophers is Tahafut al Falasifa, which has been translated by the medieval Latin translator as Destructio Philosophorum. The name of Averroës’ book is Tahafut al Tahafut, which is rendered as Destructio Destructionis (or destructionum). This rendering is surely not exact. The word ‘Tahafut’ has been translated by modern scholars in different ways, and the title of Ghazali’s book has been given as the breakdown, the disintegration, or the incoherence, of the philosophers. The exact title of Averroës’ book would be The Incoherence of the Incoherence.

In the Revue des Deux Mondes there was an article published in 1895 by Ferdinand Brunetiere, ‘La Banqueroute de la Science’, in which he tried to show that the solutions by science, and especially by biology, of fundamental problems, solutions which were in opposition to the dogmas taught by the Church, were primitive and unreasonable. Science had promised us to eliminate mystery, but, Brunetiere said, not only had it not removed it but we saw clearly that it would never do so. Science had been able neither to solve, nor even to pose, the questions that mattered: those that touched the origin of man, the laws of his conduct, his future destiny. What Brunetiere tried to do, to defend Faith by showing up the audacity of Science in its attempt to solve ultimate problems, is exactly the same as Ghazali tried to do in relation to the pretensions of the philosophers of his time who, having based themselves on reason alone, tried to solve all the problems concerning God and the world. Therefore a suitable title for his book might perhaps be ‘The Bankruptcy of Philosophy’.

In the introduction to his book Ghazali says that a group of people hearing the famous names Socrates, Hippocrates, Plato, and Aristotle, and knowing what they had attained in such sciences as Geometry, Logic, and Physics, have left the religion of their fathers in which they were brought up to follow the philosophers. The theories of the philosophers are many, but Ghazali will attack only one, the greatest, Aristotle; Aristotle, of whom it is said that he refuted all his predecessors, even Plato, excusing himself by saying ‘amicus Plato, amica veritas, sed magis amica veritas’. I may add that this well-known saying, which is a variant of a passage in Plato’s Phaedo and in Aristotle’s Nicomachean Ethics, is found in this form first in Arabic. One of the first European authors who has it in this form is Cervantes (Don Quijote, ii, c. 52). I quote this saying — Ghazali adds — to show that there is no surety and evidence in Philosophy. According to Ghazali, the philosophers claim for their metaphysical proofs the same evidence as is found in Mathematics and Logic. But all Philosophy is based on supposition and opinion. If Metaphysics had the same evidence as Mathematics all philosophers would agree just as well in Philosophy as in Mathematics. According to him the translators of Aristotle have often misunderstood or changed the meaning and the different texts have caused different controversies. Ghazali considers Farabi and Avicenna to be the best commentators on Aristotle in Islam, and it is their theories that he will attack.

Before entering into the heart of the matter I will say a few words about Ghazali’s remark that Metaphysics, although it claims to follow the same method as Mathematics, does not attain the same degree of evidence. Neither Aristotle nor his commentators ever asked the question whether there is any difference between the methods of Mathematics and Metaphysics (it is a significant fact that most examples of proof in the Posterior Analytics are taken from Mathematics) and why the conclusions reached by Metaphysics seem so much less convincing than those reached by Mathematics. It would seem that Metaphysics, being the basis of all knowledge and having as its subject the ultimate principles of things, should possess, according to Aristotle, the highest evidence and that God, as being the highest principle, should stand at the beginning of the system, as in Spinoza. In fact, Aristotle could not have sought God if he had not found Him. For Aristotle all necessary reasoning is deductive and exclusively based on syllogism. Reasoning — he says — and I think this is a profound and true remark — cannot go on indefinitely. You cannot go on asking for reasons infinitely, nor can you reason about a subject which is not known to you. Reason must come to a stop. There must be first principles which are immediately evident. And indeed Aristotle acknowledges their existence. When we ask, however, what these first principles are, he does not give us any answer but only points out the Laws of Thought as such. But from the Laws of Thought nothing can be deduced, as Aristotle acknowledges himself. As a matter of fact Aristotle is quite unaware of the assumption on which his system is based. He is what philosophers are wont to call nowadays a naive realist. He believes that the world which we perceive and think about with all it contains has a reality independent of our perceptions or our thoughts. But this view seems so natural to him that he is not aware that it could be doubted or that any reason might be asked for it. Now I, for my part, believe that the objectivity of a common world in which we all live and die is the necessary assumption of all reasoning and thought. I believe indeed, with Aristotle, that there are primary assumptions which cannot be deduced from other principles. All reasoning assumes the existence of an objective truth which is sought and therefore is assumed to have an independent reality of its own. Every thinking person is conscious of his own identity and the identity of his fellow beings from whom he accepts language and thoughts and to whom he can communicate his own ideas and emotions. Besides, all conceptual thought implies universality, i.e. belief in law and in objective necessity. I can only infer from Socrates being a man that he is mortal when I have assumed that the same thing (in this case man in so far as he is man) in the same conditions will always necessarily behave in the same way.

In his book Ghazali attacks the philosophers on twenty points. Except for the last two points which are only slightly touched by Averroës, Averroës follows point for point the arguments Ghazali uses and tries to refute them. Ghazali’s book is badly constructed, it is unsystematic and repetitive. If Ghazali had proceeded systematically he would have attacked first the philosophical basis of the system of the philosophers — namely their proof for the existence of God, since from God, the Highest Principle, everything else is deduced. But the first problem Ghazali mentions is the philosphers’ proof for the eternity of the world. This is the problem which Ghazali considers to be the most important and to which he allots the greatest space, almost a quarter of his book. He starts by saying rather arbitrarily that the philosophers have four arguments, but, in discussing them, he mixes them up and the whole discussion is complicated by the fact that he gives the philosophical arguments and theological counter arguments in such an involved way that the trend is sometimes hard to follow. He says, for instance, page 3, that to the first arguments of the philosophers there are two objections. The first objection he gives on this page, but the second, after long controversy between the philosophers and theologians, on page 32. I will not follow here Ghazali and Averroës point for point in their discussions but will give rather the substance of their principal arguments (for a detailed discussion I refer to my notes).

The theory of the eternity of the world is an Aristotelian one. Aristotle was, as he says himself, the first thinker who affirmed that the world in which we live, the universe as an orderly whole, a cosmos, is eternal. All the philosophers before him believed that the world had come into being either from some primitive matter or after a number of other worlds. At the same time Aristotle believes in the finitude of causes. For him it is impossible that movement should have started or can continue by itself. There must be a principle from which all movement derives. Movement, however, by itself is eternal. It seems to me that this whole conception is untenable. If the world is eternal there will be an infinite series of causes and an infinite series of movers; there will be an infinite series, for instance, of fathers and sons, of birds and eggs (the example of the bird and egg is first mentioned in ‘Censorinus, De die natali, where he discusses the Peripatetic theory of the eternity of the world), and we will never reach a first mover or cause, a first father or a first bird. Aristotle, in fact, defends the two opposite theses of Kant’s first antinomy. He holds at the same time that time and movement are infinite and that every causal series must be finite. The contradiction in Aristotle is still further accentuated in the Muslim philosophers by the fact that they see in God, not only as Aristotle did, the First Mover of the movement of the universe, but that they regard Him, under the influence of the Plotinian theory of emanation, as the Creator of the universe from whom the world emanates eternally. However, can the relation between two existing entities qua existents be regarded as a causal one? Can there be a causal relation between an eternally unchangeable God and an eternally revolving and changing world, and is it sense to speak of a creation of that which exists eternally? Besides, if the relation between the eternal God and the eternal movement of the world could be regarded as a causal relation, no prior movement could be considered the cause of a posterior movement, and sequences such as the eternal sequence of fathers and sons would not form a causal series. God would not be a first cause but the Only Cause of everything. It is the contradiction in the idea of an eternal creation which forms the chief argument of Ghazali in this book. In a later chapter, for instance, when he refutes Avicenna’s proof for God based on the Aristotelian concepts ‘necessary by itself’, i.e. logical necessity, and ‘necessary through another’, i.e. ontological necessity, in which there is the usual Aristotelian confusion of the logical with the ontological, Ghazali’s long argument can be reduced to the assertion that once the possibility of an infinite series of causes is admitted, there is no sense in positing a first cause.

The first argument is as follows. If the world had been created, there must have been something determining its existence at the moment it was created, for otherwise it would have remained in the state of pure possibility it was in before. But if there was something determining its existence, this determinant must have been determined by another determinant and so on ad infinitum, or we must accept an eternal God in whom eternally new determinations may arise. But there cannot be any new determinations in an eternal God.

The argument in this form is found in Avicenna, but its elements are Aristotelian. In Cicero’s Academics we have a fragment of one of Aristotle’s earlier and more popular writings, the lost dialogue De philosophia, in which he says that it is impossible that the world could ever have been generated. For how could there have been a new decision, that is a new decision in the mind of God, for such a magnificent work? St. Augustine knows this argument from Cicero and he too denies that God could have a novum consilium. St. Augustine is well aware of the difficulty, and he says in his De civilate dei that God has always existed, that after a certain time, without having changed His will, He created man, whom He had not wanted to create before, this is indeed a fact too profound for us. It also belongs to Aristotle’s philosophy that in all change there is a potentiality and all potentiality needs an actualizer which exists already. In the form this argument has in Avicenna it is, however, taken from a book by a late Greek Christian commentator of Aristotle, John Philoponus, De aeternitate mundi, which was directed against a book by the great Neoplatonist Proclus who had given eighteen arguments to prove the eternity of the world. Plato himself believed in the temporal creation of the world not by God Himself but by a demiurge. But later followers of Plato differed from him on this point. Amongst the post-Aristotelian schools only the Stoics assumed a periodical generation and destruction of the world. Theophrastus had already tried to refute some of the Stoic arguments for this view, and it may well be that John Philoponus made use of some Stoic sources for his defence of the temporality of the world.

The book by Proclus is lost, but John Philoponus, who as a Christian believes in the creation of the world, gives, before refuting them, the arguments given by Proclus. The book by Philoponus was translated into Arabic and many of its arguments are reproduced in the Muslim controversies about the problem (arguments for the temporal creation of the world were also given by Philoponus in a work against Aristotle’s theory of the eternity of the world, arguments which are known to us through their quotation and refutation by Simplicius in his commentary on Physics viii; one of these arguments by Philoponus was well known to the Arabs and is also reproduced by Ghazali, see note 3. 3). The argument I have mentioned is the third as given by Proclus. Philoponus’ book is extremely important for all medieval philosophy, but it has never been translated into a modern language and has never been properly studied. On the whole the importance of the commentators of Aristotle for Arabic and medieval philosophy in general has not yet been sufficiently acknowledged.

To this argument Ghazali gives the following answer, which has become the classic reply for this difficulty and which has been taken from Philoponus. One must distinguish, says Philoponus, between God’s eternally willing something and the eternity of the object of His Will, or, as St. Thomas will say later, ‘Deus voluit ab aeterno mundus esset sed non ut ab aeterno esset’. God willed, for instance, that Socrates should be born before Plato and He willed this from eternity, so that when it was time for Plato to be born it happened. It is not difficult for Averroës to refute this argument. In willing and doing something there is more than just the decision that you will do it. You can take the decision to get up tomorrow, but the actual willing to get up can be done only at the moment you do it, and there can be no delay between the cause and the effect. There must be added to the decision to get up the impulse of the will to get up. So in God there would have to be a new impulse, and it is just this newness that has to be denied. But, says Averroës, the whole basis of this argument is wrong for it assumes in God a will like a human will. Desire and will can be understood only in a being that has a need; for the Perfect Being there can be no need, there can be no choice, for when He acts He will necessarily do the best. Will in God must have another meaning than human will.

Averroës therefore does not explicitly deny that God has a will, but will should not be taken in its human sense. He has much the same conception as Plotinus, who denies that God has the power to do one of two contraries (for God will necessarily always choose the best, which implies that God necessarily will always do the best, but this in fact annuls the ideas of choice and will), and who regards the world as produced by natural necessity. Aristotle also held that for the Perfect Being no voluntary action is possible, and he regards God as in an eternal blissful state of self-contemplation. This would be a consequence of His Perfection which, for Averroës at least, involves His Omniscience. For the Perfect the drama of life is ended: nothing can be done any more, no decision can be taken any more, for decisions belong to the condition of man to whom both knowledge and ignorance are given and who can have an hypothetical knowledge of the future, knowing that on his decisions the future may depend and to whom a sure knowledge of the future is denied. But an Omniscient Being can neither act nor decide; for Him the future is irremediable like the past and cannot be changed any more by His decisions or actions. Paradoxically the Omnipotent is impotent. This notion of God as a Self-contemplating Being, however, constitutes one of the many profound contradictions in Aristotle’s system. And this profound contradiction is also found in all the works of Aristotle’s commentators. One of Aristotle’s proofs for the existence of God — and according to a recent pronouncement of the Pope, the most stringent — is the one based on movement. There cannot be an infinite series of movers; there must be a Prime Agent, a Prime Mover, God, the originator of all change and action in the universe. According to the conception of God as a Self-Contemplating Being, however, the love for God is the motive for the circular motion of Heaven. God is not the ultimate Agent, God is the ultimate Aim of desire which inspires the Heavens to action. It is Heaven which moves itself and circles round out of love for God. And in this case it is God who is passive; the impelling force, the efficient cause, the spring of all action lies in the world, lies in the souls of the stars.

Let us now return to Ghazali. We have seen that his first argument is not very convincing, but he now gives us another argument which the Muslim theologians have taken from John Philoponus and which has more strength. It runs: if you assume the world to have no beginning in time, at any moment which we can imagine an infinite series must have been ended. To give an example, every one of us is the effect of an infinite series of causes; indeed, man is the finite junction of an infinite past and an infinite future, the effect of an infinite series of causes, the cause of an infinite series of effects. But an infinite series cannot be traversed. If you stand near the bed of a river waiting for the water to arrive from an infinitely distant source you will never see it arriving, for an infinite distance cannot be passed. This is the argument given by Kant in the thesis of his first antimony. The curious fact is that the wording in Kant is almost identical with that of John Philoponus.

The answers Averroës gives are certainly not convincing. He repeats the Aristotelian dictum that what has no beginning has no end and that therefore there is never an end of time, and one can never say that at any moment an infinite time is ended: an infinite time is never ended. But this is begging the question and is surely not true, for there are certainly finite times. He denies that an infinite time involves an infinite causal series and the negation of a First Cause. The series involved is but a temporal sequence, causal by accident, since it is God who is its essential cause. Averroës also bases his answer on the Aristotelian theory that in time there is only a succession. A simultaneous infinite whole is denied by Aristotle and therefore, according to Aristotle, the world must be limited in space; but in time, according to him, there is never a whole, since the past is no longer existent and the future not yet.

But the philosophers have a convincing argument for the eternity of the world. Suppose the world had a beginning, then before the world existed there was empty time; but in an empty time, in pure emptiness, there cannot be a motive for a beginning and there could be nothing that could decide God to start His creation. This is Kant’s antithesis of his first antinomy. It is very old and is given by Aristotle, but it is already found in the pre-Socratic philosopher Parmenides. Ghazali’s answer is that God’s will is completely undetermined. His will does not depend on distinctions in outside things, but He creates the distinctions Himself. The idea of God’s creative will is of Stoic origin. According to the Neoplatonic conception God’s knowledge is creative. We know because things are; things are because God knows them. This idea of the creative knowledge of God has a very great diffusion in philosophy (just as our bodies live by the eternal spark of life transmitted to us by our ancestors, so we rekindle in our minds the thoughts of those who are no more); it is found, for instance, in St. Augustine, Thomas Aquinas, Spinoza, and Kant-who calls it intellektuelle Anschauung, intellectual intuition, and it is also used by the Muslim philosophers when it suits them. Against Ghazali’s conception, however, Averroës has the following argument: If God creates the world arbitrarily, if His Will establishes the distinctions without being determined by any reason, neither wisdom nor goodness can be attributed to Him. We have here a difficulty the Greeks had seen already. Either God is beyond the laws of thought and of morals and then He is neither good nor wise, or He Himself stands under their dominion and then He is not omnipotent.

Another argument for the eternity of the world is based on the eternity of time: God cannot have a priority to time, as the theologians affirm, because priority implies time and time implies movement. For the philosophers God’s priority to the world consists solely in His being its simultaneous cause. Both parties, however, seem to hold that God’s existence does not imply time, since He exists in timeless eternity. But in this case, what neither of the parties has seen, no causal relation between God and the world can exist at all, since all causation implies a simultaneous time.

We come now to the most important argument which shows the basic difference between the philosophical and theological systems. For Aristotle the world cannot have come to be because there is no absolute becoming. Everything that becomes comes from something. And, as a matter of fact, we all believe this. We all believe more or less unconsciously (we are not fully aware of our basic principles: a basement is always obscure) in the dictum rien ne se crée, rien ne se perd. We believe that everything that comes to be is but a development, an evolution, without being too clear about the meaning of these words (evolution means literally ‘unrolling’, and Cicero says that the procession of events out of time is like the uncoiling of a rope — quasi rudentis explicatio), and we believe that the plant lies in the seed, the future in the present. For example: when a child is born we believe it to have certain dispositions; it may have a disposition to become a musician, and when all the conditions are favourable it will become a musician. Now, according to Aristotle, becoming is nothing but the actualization of a potentiality, that is the becoming actual of a disposition. However, there is a difficulty here. It belongs to one of the little ironies of the history of philosophy that Aristotle’s philosophy is based on a concept, i.e. potentiality, that has been excluded by a law that he was the first to express consciously. For Aristotle is the first to have stated as the supreme law of thought (or is it a law of reality?) that there is no intermediary between being and non-being. But the potential, i.e. the objective possible, is such an intermediary; it is namely something which is, still is not yet. Already the Eleatics had declared that there is no becoming, either a thing is or it is not. If it is, it need not become. If it is not-out of nothing nothing becomes. Besides, there is another difficulty which the Megarians have shown.

You say that your child has a disposition to become a musician, that he can become a musician, but if he dies as a child, or when conditions are unfavourable, he cannot become a musician. He can only become one when all the conditions for his being a musician are fulfilled. But in that case it is not possibly that he will be a musician, necessarily he will be one. There is in fact no possibility of his being a musician before he actually is one. There is therefore no potentiality in nature and no becoming of things out of potencies. Things are or are not. This Megarian denial of potentiality has been taken over by the Ash‘arites, and Ghazali in this book is on the whole, although not consistently, in agreement with them. I myself regard this problem as one of the cruces of philosophy. The Ash‘arites and Ghazali believed, as the Megarians did, that things do not become and that the future does not lie in the present; every event that occurs is new and unconnected with its predecessor. The theologians believed that the world is not an independent universe, a self-subsistent system, that develops by itself, has its own laws, and can be understood by itself. They transferred the mystery of becoming to the mystery of God, who is the cause of all change in the world, and who at every moment creates the world anew. Things are or are not. God creates them and annihilates them, but they do not become out of each other, there is no passage between being and non-being. Nor is there movement, since a thing that moves is neither here nor there, since it moves — what we call movement is being at rest at different space — atoms at different time — atoms. It is the denial of potentiality, possibility in rerum natura, that Ghazali uses to refute the Aristotelian idea of an eternal matter in which the potentialities are found of everything that can or will happen. For, according to Aristotle, matter must be eternal and cannot have become, since it is, itself, the condition for all becoming.

It maybe mentioned here that the modern static theory of movement is akin to the Megarian-Ash‘arite doctrine of the denial of movement and becoming. Bertrand Russell, for instance, although he does not accept the Megarian atomic conception, but holds with Aristotle that movement and rest take place in time, not in the instant, defines movement as being at different places at different times. At the same time, although he rejects the Megarian conception of ‘jumps’, he affirms that the moving body always passes from one position to another by gradual transition. But ‘passing’ implies, just as much as ‘jumping’, something more than mere being, namely, the movement which both theories deny and the identity of the moving body.

On the idea of possibility another argument for the eternity of the world is based. It is affirmed that if the world had been created an infinite number of possibilities of its creation, that is, an eternal duration of its possibility, would have preceded it. But nothing possible can be eternal, since everything possible must be realized. The idea that everything possible has to be realized is found in Aristotle himself, who says that if there could be an eternal possible that were not realized, it would be impossible, not possible, since the impossible is that which will never be realized. Aristotle does not see that this definition is contrary to the basic idea of his own philosophy — the reality of a possibility which may or may not become real — and that by declaring that the possible will have to happen he reduces it to a necessity, and by admitting that everything that happens had to happen he denies that the possibility of its not happening could precede it, i.e. he accepts, in fact, the Megarian conception of possibility which he himself had tried to refute. Averroës, who agrees with his master on this point, is not aware either of the implication of the definition. On the other hand, the Ash‘arites, notwithstanding their denial of potentiality, maintain that for God everything is possible, a theory which implies objective possibility (the same inconsistency was committed by the Stoics). Both philosophers and theologians, indeed, hold about this difficult problem contradictory theories, and it is therefore not astonishing that Ghazali’s and Averroës’ discussion about it is full of confusion (for the details I refer to my notes).

In the second chapter Ghazali treats the problem of the incorruptibility of the world. As Ghazali says himself; the problem of the incorruptibility of the world is essentially the same as that of its being uncreated and the same arguments can be brought forward. Still, there is less opposition amongst the theologians about its incorruptibility than about its being uncreated. Some of the Mu‘tazilites argued, just as Thomas Aquinas was to do later, that we can only know through the Divine Law that this world of ours will end and there is no rational proof for its annihilation. Just as a series of numbers needs a first term but no final term, the beginning of the world does not imply its end. However, the orthodox view is that the annihilation of the world, including Heaven and Hell, is in God’s power, although this will not happen. Still, in the corruptibility of the world there is a new difficulty for the theologians. If God destroys the world He causes ‘nothingness’, that is, His act is related to ‘nothing’. But can an act be related to ‘nothing’? The question as it is posed seems to rest on a confusion between action and effect but its deeper sense would be to establish the nature of God’s action and the process by which His creative and annihilating power exercises itself. As there cannot be any analogy with the physical process through which our human will performs its function, the mystery of His creative and annihilating action cannot be solved and the naive answers the theologians give satisfy neither Averroës nor Ghazali himself. Averroës argues that there is no essential difference between production and destruction and, in agreement with Aristotle, he affirms that there are three principles for them: form, matter, and privation. When a thing becomes, its form arises and its privation disappears; when it is destroyed its privation arises and its form disappears, but the substratum of this process, matter, remains eternally. I have criticized this theory in my notes and will only mention here that for Aristotle and Averroës this process of production and destruction is eternal, circular, and reversible. Things, however, do not revolve in an eternal cycle, nor is there an eternal return as the Stoics and Nietzsche held. Inexorably the past is gone. Every ‘now’ is new. Every flower in the field has never been, the up-torn trees are not rooted again. ‘Thou’ll come no more, Never, never, never, never, never!’ Besides, Averroës, holding as he does that the world is eternally produced out of nothing, is inconsistent in regarding with Aristotle production and destruction as correlatives.

In the third chapter Ghazali maintains that the terms acting and agent are falsely applied to God by the philosophers. Acting, according to him, can be said only of a person having will and choice. When you say that fire burns, there is here a causal relation, if you like, but this implies nothing but a sequence in time, just as Hume will affirm later. So when the philosophers say that God’s acting is like the fire’s burning or the sun’s heating, since God acts by natural necessity, they deny, according to Ghazali, His action altogether. Real causation can only be affirmed of a willing conscious being. The interesting point in this discussion is that, according to the Ash‘arites and Ghazali, there is no causation in this world at all, there is only one extra-mundane cause which is God. Even our acts which depend on our will and choice are not, according to the Ash‘arites, truly performed by ourselves. We are only the instruments, and the real agent is God. But if this is true, how can we say that action and causation depend on will and choice? How can we come to the idea of any causal action in God depending on His Will if we deny generally that there is a causal relation between will and action? The same contradiction is found in modern philosophy in Mach. Mach holds that to speak of causation or action in material things — so to say that fire burns — is a kind of fetishism or animism, i.e. that we project our will and our actions into physical lifeless things. However, at the same time he, as a follower of Hume, says that causation, even in acts caused by will, is nothing but a temporal sequence of events. He denies causation even in voluntary actions. Therefore it would follow that the relation of willing and acting is not different from the relation of fire and burning and that there cannot be any question of fetishism or animism. According to such a theory there is no action at all in the universe but only a sequence of events.

Then, after a second argument by which Ghazali sets out to show that an eternal production and creation are contradictions in terms, since production and creation imply the generation of something after its non-existence, he directs a third argument against the Neoplatonic theory, held by the philosophers, of the emanation of the world from God’s absolute Oneness.

Plotinus’ conception of God is prompted by the problem of plurality and relation. All duality implies a relation, and every relation establishes a new unity which is not the simple addition of its terms (since every whole is more than its parts) and violates therefore the supreme law of thought that a thing is what it is and nothing else. Just as the line is more than its points, the stone more than its elements, the organism transcending its members, man, notwithstanding the plurality of his faculties, an identical personality, so the world is an organized well-ordered system surpassing the multitude of the unities it encloses. According to Plotinus the Force binding the plurality into unity and the plurality of unities into the all-containing unit of the Universe is the Archetype of unity, the ultimate, primordial Monad, God, unattainable in His supreme Simplicity even for thought. For all thought is relational, knitting together in the undefinable unity of a judgement a subject and a predicate. But in God’s absolute and highest Unity there is no plurality that can be joined, since all joining needs a superior joining unit. Thus God must be the One and the Lone, having no attribute, no genus, no species, no universal that He can share with any creatures of the world. Even existence can be only referred to Him when it expresses not an attribute, but His very Essence. But then there is no bridge leading from the stable stillness of His Unity to the changing and varied multiplicity of the world; all relation between Him and the world is severed. If the One is the truly rational, God’s rationality can be obtained only by regarding His relation to the world as irrational, and all statements about Him will be inconsistent with the initial thesis. And if God is unattainable for thought, the very affirmation of this will be self-contradictory.

Now, the philosophers in Islam hold with Plotinus that although absolutely positive statements are not admissible about God, the positive statements made by them can be all reduced to negative affirmations (with the sole exception, according to Averroës, of His possessing intellect) and to certain relative statements, for neither negations nor external relations add anything to His essence.

In this and several following chapters Ghazali attacks the philosophers from two sides: by showing up the inanity of the Plotinian conception of God as pure unity, and by exposing their inconsistency in attributing to Him definite qualities and regarding Him as the source of the world of variety and plurality.

The infinite variety and plurality of the world does not derive directly from God according to the philosophers in Islam, who combine Aristotle’s astronomical view of animate planets circling round in their spheres with the Neoplatonic theory of emanation, and introduce into the Aristotelian framework Proclus’ conception of a triadic process, but through a series of immaterial mediators. From God’s single act — for they with Aristotle regard God as the First Agent — only a single effect follows, but this single effect, the supramundane Intellect, develops in itself a threefoldness through which it can exercise a threefold action. Ghazali objects in a long discussion that if God’s eternal action is unique and constant, only one single effect in which no plurality can be admitted will follow (a similar objection can be directed against Aristotle, who cannot explain how the plurality and variety of transitory movements can follow from one single constant movement). The plurality of the world according to Ghazali cannot be explained through a series of mediators. Averroës, who sometimes does not seem very sure of the validity of mediate emanation, is rather evasive in his answer on this point.

In a series of rather intricate discussions which I have tried to elucidate in my notes, Ghazali endeavours to show that the proofs of the philosophers for God’s uniqueness, for their denial of His attributes, for their claims that nothing can share with Him His genus and species, that He is pure existence which stands in no relation to an essence, and that He is incorporeal, are all vain. The leading idea of the philosophers that all plurality needs a prior joining principle, Ghazali rejects, while Averroës defends it. Why — so Ghazali asks, for instance — since the essence in temporal things is not the cause of their existence, should this not be the case in the Eternal? Or why should body, although it is composite according to the philosophers, not be the First Cause, especially as they assume an eternal body, since it is not impossible to suppose a compound without a composing principle? From the incorporeality of God, the First Principle, Avicenna had tried to infer, through the disjunction that everything is either matter or intellect, that He is intellect (since the philosophers in Islam hold with Aristotle and in opposition to Plotinus that God possesses self-consciousness). Ghazali does not admit this disjunction and, besides, argues with Plotinus that self-consciousness implies a subject and an object, and therefore would impede the philosophers’ thesis of God’s absolute unity.

The Muslim philosophers, following Aristotle’s Neoplatonic commentators, affirm that God’s self-knowledge implies His knowledge of all universals (a line of thought followed, for instance, by Thomas Aquinas and some moderns like Brentano). In man this knowledge forms a plurality, in God it is unified. Avicenna subscribes to the Qur’anic words that no particle in Heaven or Earth escapes God’s knowledge, but he holds, as Porphyry had done before, that God can know the particular things only in a universal way, whatever this means. Ghazali takes it to mean that God, according to Avicenna, must be ignorant of individuals, a most heretical theory. For Averroës God’s knowledge is neither universal nor particular, but transcending both, in a way unintelligible to the human mind.

One thing, however, God cannot know according to Avicenna (and he agrees here with Plato’s Parmenides) and that is the passing of time, for in the Eternal no relation is possible to the fleeting ‘now’. There are two aspects of time: the sequence of anteriority and posteriority which remains fixed for ever, and the eternal flow of the future through the present into the past. It will be eternally true that I was healthy before I sickened and God can know its eternal truth. But in God’s timeless eternity there can be no ‘now’ simultaneous with the trembling present in which we humans live and change and die, there is no ‘now’ in God’s eternity in which He can know that I am sickening now. In God’s eternal stillness the fleeting facts and truths of human experience can find no rest. Ghazali objects, erroneously, I think, that a change in the object of thought need not imply a change in the subject of consciousness.

In another chapter Ghazali refutes the philosophers’ proof that Heaven is animated. He does not deny its possibility, but declares that the arguments given are insufficient. He discusses also the view that the heavens move out of love for God and out of desire to assimilate themselves to Him, and he asks the pertinent question — already posed by Theophrastus in his Metaphysics, but which scandalizes Averroës by its prosaicness — why it is meritorious for them to circle round eternally and whether eternal rest would not be more appropriate for them in their desire to assimilate themselves to God’s eternal stability.

In the last chapter of this part Ghazali examines the philosophers’ symbolical interpretation of the Qur’anic entities ‘The Pen’ and ‘The Tablet’ and their theories about dreams and prophecy. It is interesting to note that, although he refutes them here, he largely adopts them in his own Vivification of Theology. [?]

In the last part of his book Ghazali treats the natural sciences. He enumerates them and declares that there is no objection to them according to religion except on four points. The first is that there exists a logical nexus between cause and effect; the second, the selfsubsistent spirituality of the soul; the third, the immortality of this subsistent soul; the fourth, the denial of bodily resurrection. The first, that there exists between cause and effect a logical necessity, has to be contested according to Ghazali, because by denying it the possibility of miracles can be maintained. The philosophers do not deny absolutely the possibility of miracles. Muhammad himself did not claim to perform any miracles and Hugo Grotius tried to prove the superiority of Christianity over Islam by saying ‘Mahumetis se missum ait non cum miraculis sed cum armis’. In later times, however, Muhammad’s followers ascribed to him the most fantastic miracles, for instance the cleavage of the moon and his ascension to Heaven. These extravagant miracles are not accepted by the philosophers. Their theory of the possibility of miracles is based on the Stoic-Neoplatonic theory of ‘Sympathia’, which is that all parts of the world are in intimate contact and related. In a little treatise of Plutarch it is shown how bodily phenomena are influenced by suggestion, by emotion and emotional states, and it is claimed by him, and later also by Plotinus, that the emotions one experiences cannot only influence one’s own body but also other bodies, and that one’s soul can exercise an influence on other bodies without the intermediary of any bodily action. The phenomena of telepathy, for instance the fascination which a snake has on other animals, they explained in this way. Amulets and talismans can receive through psychological influences certain powers which can be realized later. This explanation of occult phenomena, which is found in Avicenna’s Psychology, a book translated in the Middle Ages, has been widely accepted (for instance, by Ghazali himself in his Vivification of Theology), and is found in Thomas Aquinas and most of the writers about the occult in the Renaissance, for instance Heinricus Cornelius Agrippa, Paracelsus, and Cardanus. It may be mentioned here that Avicenna gives as an example of the power of suggestion that a man will go calmly over a plank when it is on the ground, whereas he will hesitate if the plank be across an abyss. This famous example is found in Pascal’s Pensées, and the well-known modern healer, Coué, takes it as his chief proof for the power of suggestion. Pascal has taken it from Montaigne, Montaigne has borrowed it from his contemporary the great doctor Pietro Bairo, who himself has a lengthy quotation from the Psychology of Avicenna. Robert Burton in his Anatomy of Melancholy also mentions it. In the Middle Ages this example is found in Thomas Aquinas. Now the philosophers limit the possibility of miracles only to those that can be explained by the power of the mind over physical objects; for instance, they would regard it as possible that a prophet might cause rain to fall or an earthquake to take place, but they refuse to accept the more extravagant miracles I have mentioned as authentic.

The theologians, however, base their theory of miracles on a denial of natural law. The Megarian-Ash‘arite denial of potentiality already implies the denial of natural law. According to this conception there is neither necessity nor possibility in rerum natura, they are or they are not, there is no nexus between the phenomena. But the Greek Sceptics also deny the rational relation between cause and effect, and it is this Greek Sceptical theory which the Ash‘arites have copied, as we can see by their examples. The theory that there is no necessary relation between cause and effect is found, for instance, in Galen. Fire burns but there is, according to the Greek Sceptics, no necessary relation between fire and burning. Through seeing this happen many times we assume that it will happen also in the future, but there is no necessity, no absolute certainty. This Sceptical theory is quasi-identical with the theory of Hume and is based on the same assumptions, that all knowledge is given through sense-impression; and since the idea of causation cannot be derived from sense experience it is denied altogether. According to the theory of the theologians, God who creates and re-creates the universe continually follows a certain habit in His creation. But He can do anything He desires, everything is possible for Him except the logically impossible; therefore all logically possible miracles are allowed. One might say that, for the theologians, all nature is miraculous and all miracles are natural. Averroës asks a good question: What is really meant by habit, is it a habit in man or in nature? I do not know how Hume would answer this question. For if causation is a habit in man, what makes it possible that such a habit can be formed? What is the objective counterpart of these habits? There is another question which has been asked by the Greek opponents of this theory, but which is not mentioned by Averroës: How many times must such a sequence be observed before such a habit can be formed? There is yet another question that might be asked: Since we cannot act before such a habit is formed — for action implies causation — what are we doing until then? What, even, is the meaning of ‘I act’ and ‘I do’? If there is nothing in the world but a sequence of events, the very word ‘activity’ will have no sense, and it would seem that we would be doomed to an eternal passivity. Averroës’ answer to this denial of natural law is that universals themselves imply already the idea of necessity and law. I think this answer is correct. When we speak, for instance, of wood or stone, we express by those words an hypothetical necessity, that is, we mean a certain object, which in such-and-such circumstances will necessarily behave in a certain way that the behaviour of wood, for example, is based on its nature, that is, on the potentialities it has.

I may remark here that it seems to me probable that Nicholas of Autrecourt, ‘the medieval Hume’, was influenced by Ghazali’s Ash‘arite theories. He denies in the same way as Ghazali the logical connexion between cause and effect: ‘ex eo quod aliqua res est cognita esse, non potest evidenter evidentia reducta in primum principium vel in certitudinem primi principii inferri, quod alia res sit’ (cf. Lappe, ‘Nicolaus von Autrecourt’, Beitr. z. Gesch. d. Phil. d. M. B.vi, H.2, p. 11); he gives the same example of ignis and stupa, he seems to hold also the Ash‘arite thesis of God as the sole cause of all action (cf. op. cit., p. 24), and he quotes in one place Ghazali’s Metaphysics (cf. N. of Autrecourt, ‘Exigit ordo executionis’, in Mediaeval Studies, vol. i, ed. by J. Reginald O’Donnell, Toronto, 1931, p. 2o8). Now Nicholas’s works were burnt during his lifetime in Paris in 1347, whereas the Latin translation of the Tahafut al Tahafut by Calo Calonymus was terminated in Arles in 1328.

The second point Ghazali wants to refute are the proofs for the substantiality and the spirituality of the soul as given by the philosophers. He himself does not affirm that the soul is material, and as a matter of fact he holds, in other books, the contrary opinion, but the Ash‘arites largely adopted the Stoic materialism. The ten arguments of the philosophers for the spirituality of the soul derive all from arguments given by the Greeks. It would seem to me that Ghazali’s arguments for the soul’s materiality may be based on the Stoic answers (which have not come down to us) against the proofs of Aristotle and the later Platonists for the immateriality of the soul. There is in the whole discussion a certain confusion, partly based on the ambiguity of the word ‘soul’. The term ‘soul’ both in Greek and Arabic can also mean ‘life’. Plants and animals have a ‘soul’. However, it is not affirmed by Aristotle that life in plants and animals is a spiritual principle. ‘Soul’ is also used for the rational part, the thinking part, of our consciousness. It is only this thinking part, according to Aristotle, that is not related to or bound up with matter; sensation and imagination are localized in the body, and it is only part of our thinking soul that seems to possess eternity or to be immortal. Now, most of the ten arguments derive from Aristotle and mean only to prove that the thinking part of our soul is incorporeal. Still the Muslim philosophers affirm with Plato and Plotinus that the whole soul is spiritual and incorruptible, and that the soul is a substance independent of the body, although at the same time they adopt Aristotle’s physiological explanations of all the non-rational functions of the soul and accept Aristotle’s definition of the ‘soul’ as the first entelechy of an organic body. On the other hand, the Muslim philosophers do not admit the Platonic theory of the pre-existence of the soul. Aristotle’s conception of a material and transitory element in the soul and an immaterial and immortal element destroys all possibility of considering human personality as a unity. Although he reproaches Plato with regarding the human soul as a plurality, the same reproach can be applied to himself. Neither the Greek nor the Muslim philosophers have ever been able to uphold a theory that does justice to the individuality of the human personality. That it is my undefinable ego that perceives, represents, wills, and thinks, the mysterious fact of the uniqueness of my personality, has never been apprehended by them. It is true that there is in Aristotle’s psychology a faint conception of a functional theory of our conscious life, but he is unable to harmonize this with his psycho-physiological notions.

I have discussed in my notes the ten arguments and will mention here only two because of their importance. Ghazali gives one of these arguments in the following form: How can man’s identity be attributed to body with all its accidents? For bodies are continually in dissolution and nutrition replaces what is dissolved, so that when we see a child, after separation from its mother’s womb, fall ill a few times, become thin and then fat again, and grow up, we may safely say that after forty years no particle remains of what there was when its mother was delivered of it. Indeed, the child began its existence out of parts of the sperm alone, but nothing of the particles of the sperm remains in it; no, all this is dissolved and has changed into something else and then this body has become another. Still we say that the identical man remains and his notions remain with him from the beginning of his youth although all bodily parts have changed, and this shows that the soul has an existence outside the body and that the body is its organ. Now the first part of this argument, that all things are in a state of flux and that of the bodily life of man no part remains identical, is textually found in Montaigne’s Apology of Raymond de Sebond. Montaigne has taken it from Plutarch, and the Arabic philosophers may have borrowed it from the same source from which Plutarch has taken it. The argument of the philosophers that matter is evanescent, but the soul a stable identity, which is also given by the Christian philosopher Nemesius in his De natura hominis (a book translated into Arabic), who ascribes it to Ammonius Saccas and Numenius, is basically Platonic and Neoplatonic, and strangely enough, although he refutes it here, it is adduced by Ghazali himself in his Vivification of Theology. Socrates says in the Platonic dialogue Cratylus: ‘Can we truly say that there is knowledge, Cratylus, if all things are continually changing and nothing remains? For knowledge cannot continue unless it remains and keeps its identity. But if knowledge changes its very essence, it will lose at once its identity and there will be no knowledge.’ Plotinus (Enn. iv. 7. 3) argues that matter, in its continual changing, cannot explain the identity of the soul. And he says in a beautiful passage (Enn. iv. 7. 10) the idea of which Avicenna has copied:

‘One should contemplate the nature of everything in its purity, since what is added is ever an obstacle to its knowledge. Contemplate therefore the soul in its abstraction or rather let him who makes this abstraction contemplate himself in this state and he will know that he is immortal when he will see in himself the purity of the intellect, for he will see his intellect contemplate nothing sensible, nothing mortal, but apprehending the eternal through the eternal.’

This passage bears some relation to Descartes’s dictum cogito ergo sum, but whereas Plotinus affirms the self-consciousness of a stable identity, Descartes states only that every thought has a subject, an ego. Neither the one, nor the other shows that this subject is my ego in the sense of my undefinable unique personality, my awareness who I am: that I am, for instance, John and not Peter, my consciousness of the continuity of my identity from birth to death, my knowledge that at the same time I am master and slave of an identical body, whatever the changes may be in that body, and that as long as I live I am a unique and an identical whole of body and soul. Plautus’ Sosia, who was not a philosopher, expresses himself (Amphitruo, line 447) in almost the same way as Descartes — ‘sed quom cogito, equidem certo idem sum qui fui semper’ — but the introduction of the words semper and idem renders the statement fallacious; from mere consciousness the lasting identity of my personality cannot be inferred.

Ghazali answers this point by saying that animals and plants also, notwithstanding that their matter is continually changing, preserve their identity, although nobody believes that this identity is based on a spiritual principle. Averroës regards this objection as justified.

The second argument is based on the theory of universals. Since thought apprehends universals which are not in a particular place and have no individuality, they cannot be material, since everything material is individual and is in space. Against this theory of universals Ghazali develops, under Stoic influence, his nominalistic theory which is probably the theory held by the Ash‘arites in general. This theory is quasi-identical with Berkeley’s nominalistic conception and springs from the same assumption that thinking is nothing but the having of images. By a strange coincidence both Ghazali and Berkeley give the example of a hand: when we have an idea of a hand as a universal, what really happens is that we have a representation of a particular hand, since there are no universals. But this particular hand is capable of representing for us any possible hand, just as much a big black hand as a small white one. The fallacy of the theory lies, of course, in the word ‘representing’, which as a matter of fact assumes what it tended to deny, namely, that we can think of a hand in general which has neither a particular shape, nor a particular colour, nor is localized in space.

The next point Ghazali tries to refute is the argument of the philosophers for the immortality of the soul. According to the philosophers, the fact that it is a substance independent of a body and is immaterial shows that a corruption of the body cannot affect it. This, as a matter of fact, is a truism, since the meaning of substantiality and immateriality for the philosophers implies already the idea of eternity. On the other hand, if the soul is the form of the body, as is also affirmed by them, it can only exist with its matter and the mortality of its body would imply its own mortality, as Ghazali rightly points out. The Arabic philosophers through their combination of Platonism and Aristotelianism hold, indeed, at the same time three theories inconsistent with each other, about the relation of body and soul: that the soul is the form of the body, that the soul is a substance, subsistent by itself and immortal, and that the soul after death takes a pneumatic body (a theory already found in Porphyry). Besides, their denial of the Platonic idea of pre-existence of the soul vitiates their statement that the soul is a substance, subsistent by itself, that is, eternal, ungenerated, and incorruptible. Although Averroës in his whole book tries to come as near to the Aristotelian conception of the soul as possible, in this chapter he seems to adopt the eschatology of the late Greek authors. He allows to the souls of the dead a pneumatic body and believes that they exist somewhere in the sphere of the moon. He also accepts the theory of the Djinn, the equivalent of the Greek Daimones. What he rejects, and what the philosophers generally reject, is the resurrection of the flesh.

In his last chapter Averroës summarizes his views about religion. There are three possible views. A Sceptical view that religion is opium for the people, held by certain Greek rationalists; the view that religion expresses Absolute Truth; and the intermediate view, held by Averroës, that the religious conceptions are the symbols of a higher philosophical truth, symbols which have to be taken for reality itself by the non-philosophers. For the unphilosophical, however, they are binding, since the sanctity of the State depends on them.