When considering the theology of GÚaza@l^ two related questions immediately arise. The first is his attitude towards Islamic speculative/dialectical theology (kala@m). The second is the extent to which he is committed to one of its main schools, the Ash¿arite. GÚaza@l^ was trained both as a Shafi¿ite lawyer and as an Ash¿arite theologian and had as his teacher none other than the prominent Ash¿arite theologian and Shafi¿ite lawyer, Ema@m-al-H®aramayn Abu'l-Ma¿a@l^ Z˜^a@÷-al-D^n ¿Abd-al-Malek Jovayn^ (d. 478/1085). He devoted two works to Ash¿arite kala@m and considered his refutation of the philosophers, Taha@fot al-fal@a@sefa (The incoherence of the philosophers) as belonging to the genre of kala@m, even though his declared task in this work is simply to refute the Islamic philosophers, not to develop any specific doctrinal position. At the same time, GÚaza@l^ criticized the discipline of kala@m in several of his writings, for example, in his major work, the voluminous Eháya@÷ ¿olu@m al-d^n, in his autobiography, al-Monqedò men al-zµala@l and in the last book he wrote shortly before his death, Elja@m al-¿awa@mm ¿an ¿elm al-kala@m.

GÚaza@l^ argued that the main role of kala@m is the preservation (háefzá) and guardianship (háera@sa) of true religious belief (¿aq^da), that is, traditional belief guided by the customary practice (sonna) of the Prophet. Kala@m's task is corrective: to correct distortions in the exposition of belief by the heretical innovators (ahl-al-beda÷). It is intended to persuade the few, those sincere doubters who are intellectually capable of following its arguments. For this persuasive reason, each region should have a theologian (motakallem). The "commonality," however, must not be exposed to kala@m. For GÚaza@l^, kala@m is not an end in itself. It is an error, he maintains, to think that practicing the discipline of kala@m constitutes what is experientially religious.

Nonetheless, this does not mean that for GÚaza@l^ the principles of Ash¿arite kala@m, that is, its articles of faith, are not true and that he is not committed to them; but it is one thing to expound them and argue for their validity and another to experience what these principles mean. The true meaning of these principles is not attained through rational argument, but through direct mystical experience, moka@æafa. GÚaza@l^ devotes one of the books of his major mystical work, the Eháya@÷, to an exposition of Ash¿arite theology. The book is entitled Qawa@¿ed al-¿aqa@÷ed. Its Ash¿arism blends, though not always conspicuously, with the book's mysticism. His main Ash¿arite work is the Eqtesáa@d fi'l-e¿teqa@d, where the exposition is more detailed and sustained. In the Eqtesáa@d, he affirms the Ash¿arite doctrine of a material soul. In the Eháya@÷, while his language sometimes suggests an inclination towards a non-material doctrine of the soul, the theological world view which GÚaza@l^ develops remains basically Ash¿arite. He expands on Ash¿arism, appropriating for it philosophical ideas that derive largely from Avicenna. These ideas are reinterpreted so as to accord with Ash¿arism.

The cornerstone of Ash¿arism is its doctrine of the divine attributes, to which GÚaza@l^ fully subscribes and elaborates. These attributes are not identical with the divine essence, but are "additional" to it. This distinction is important. If the attributes were identical with the divine essence, then the divine act would proceed as the necessary consequence of the divine essence or nature. This means a negation of the divine will, since then God would have no choice but to act, and act in the one way necessitated by His essence. Since His essence is eternal, His necessitated act would be eternal. The world, the necessitated effect of His eternal act, would thus be eternal. It would not be an existent temporally originated by God. It would not be a real creation.

The eternal attributes are coeternal with the divine essence. They have a special relation to each other. For example, without the attribute of life there can be no attribute of knowledge and without the attribute of knowledge there can be not attribute of will. Hence the attribute of life is a necessary condition for the attribute of knowledge, and the attribute of knowledge, a necessary condition for the attribute of will. But this necessary condition does not entail that the relation between these attributes is causal. None of the coeternal divine attributes is a cause of the other.

Whatever God eternally wills must come to be. In this sense it is necessary. But it is not necessary because it is the consequence of the divine nature. God could have chosen not to create at all. But "once," so to speak (GÚaza@l^ insists that human language regarding the divine voluntary act is metaphorical), He wills something, this something must come to be. It comes to be through the causal action of the attribute of divine power. This eternal attribute, which again is "additional" to the divine essence, is one and pervasive. By this GÚaza@l^ means (a) that this attribute does not consist of multiplicity of powers and (b) that it is the direct cause of each and every created existent and event. The world and all the events therein are thus the direct effect of this one cause. For GÚaza@l^ the world consists of contingent atoms (substances) and accidents. These are created ex nihilo, combined to form bodies and sustained in temporally finite spans of existence by the direct action of divine power. As GÚaza@l^ puts it: "…all temporal events, their substances and accidents, those occurring in the entities of the animate and the inanimate, come about through the power of God, exalted be He. He alone holds the sole prerogative of inventing them. No created thing comes about through another [created thing]. Rather, all come about through [divine] power " (Eqtesáa@d, p. 99).

In several other places in the Eqtesáa@d, GÚaza@l^ reaffirms unequivocally his position that there is only one direct cause, namely divine power, for all created things and for all temporal events. Causal efficacy resides in divine power alone. There is no real evidence in his subsequent writings to indicate that he ever departed from upholding such a view. But this doctrine of the "pervasiveness of [divine] power" (¿omu@m al-qodra) raised for GÚaza@l^ two questions which he strove to answer. The first has to do with scientific knowledge. If there are no natural causes, how is it possible for us to make scientific inferences from what is ordinarily regarded as natural cause and effect? The second is the question of the human act. Is it also created by divine power, and if so, how, in the realm of moral action, are humans responsible for their acts?

Regarding the first question, GÚaza@l^ does not deny that the events in this world have an order which we habitually regard as constituting causes and effects. But the connections between the habitually regarded causes and effects are not necessary. These habitual causes and effects follow sequences that parallel Avicenna's conception. They behave as though they are real causes and their effects, allowing us ordinarily to speak of them as causes and effects and to draw from them scientific inferences. But in reality they are not real causes and their effects. They do, however, follow an order, invariable for the most part, ordained by the divine will. This order in itself is not necessary. Its disruption is hence possible. The divine will which decrees this order also decrees its disruption at certain times in history. The disruption constitutes the occurrence of the miracle. When such an occurrence takes place, God removes from us knowledge of nature's uniformity, creating for us the knowledge of the miracle.

Turning to the question of the human act in relation to divine power, GÚaza@l^'s position is Ash¿arite. It has to be understood in terms of the Ash¿arite criticism of the doctrine of divine justice espoused by the earlier Mu¿tazilite school of kala@m. Divine justice, according to the Mu¿tazilites, entails man's freedom of the will. Man chooses and "creates" those acts for which he is rewarded or punished in the hereafter. The Mu¿tazilites held that reason discerns the moral value of acts. These values hence are objective qualities that inhere in the acts. An act is not good simply because God commands it. Rather, it is precisely because an act is in itself good that God commands it.

GÚaza@l^ rejects the Mu¿tazilite doctrine of the intrinsic value of the moral act. It imposes a limit on divine power since the divine act would be conditioned by such intrinsic objective values of acts. In themselves, he argues, acts are morally neutral. They are good or bad simply because God either commands or prohibits them. The just act is the act which God performs or commands. Hence, if God were to torture an innocent child in the hereafter, this would be good and just. We know that He does not perform such acts because He has so indicated, and God does not utter any falsehood.

Turning to the Mu¿tazilite doctrine of the freedom of the will, again for GÚaza@l^, it imposes a limitation on divine power. Creation belongs only to God. It is blasphemy to maintain that there are acts which man creates. All human acts, for GÚaza@l^, are the creation of divine power. This brings us to GÚaza@l^'s full endorsement of the Ash¿arite doctrine of acquisition (kasb), a doctrine which he argues for at length in the Eqtesáa@d and reaffirms in the Eháya@÷.

GÚaza@l^ argues unequivocally (a) that "power" in all animate creatures is created directly by divine power and (b) that there is created with it the object of power (maqdu@r) ordinarily, but erroneously, regarded as the effect of created power. The human act is created for the human by divine power. The human "acquires" it as well as the object ordinarily regarded as produced by the human act. In other words, whatever humans are said to deliberately perform is in reality created on their behalf by divine power. Divine power creates human power after it has created human life, knowledge, and will. Created power, moreover, never precedes the created act. It is created with it.

To the common objection that this leads to utter determinism (jabr), GÚaza@l^ responds that with the determinists (mojbera) there can be no distinction between the spasmodic movement and the movement over which we experience power. For created power and the movement that accompanies it are things we in fact experience. The Mu¿tazilite belief that we create our own acts cannot be true because to create something is to have knowledge of all its consequential details. We never can have such knowledge. Bees have no mathematical knowledge. Yet, their hexagonal houses are built to perfection. This is created for them by divine power. The bees are simply the locus of divine action. So are humans. But if will, power and act are all the direct creation of divine power, how can humans be morally responsible? GÚaza@l^ does not really answer this question. His language suggests that we are here in the realm of mystery. In the Eháya@÷, he indicates that the manner in which the eternal divine power is the cause of each and every existent and happening, including the human act, can only be understood by those well grounded in knowledge (al-ra@sekòu@n fi'l-¿elm) through kaæf, direct mystical revelation. For in this experience, the mystic sees that God is the only reality; all things then fall into place.

Bibliography: Texts: Eháya@÷ ¿olu@m al-d^n, 4 vols., Cairo, 1377/1957 (there are various printings of this work, but thus far, no critical edition); tr. G. H. Bousquet et al. as Ih'ya ¿ouloûm ed-dîn; ou, Vivification des sciences de la foi: analyse et index, Paris, 1955. Elja@m al-¿awa@mm ¿an ¿elm al-kala@m, ed. M. Bag@da@d^, Cairo, 1985. al-Eqtesáa@d fi'l-e¿teqa@d, ed. I. S. Çubukçu and H. Atay, Ankara, 1962; ed. ¿A. H®amaw^, Damascus, 1999. Al-Maqsáad al-asna@ f^ asma@÷ Alla@h al-háosna@, ed. F. A. Shehadi, Beirut, 1982; tr. R. Ch. Stade, as The Attributes of Divine Perfection, Jeddah, 1989; tr. D. B. Burrell and N. Daher as The Ninety-nine Beautiful Names of God, Cambridge, U.K., 1992. Al-Monqedò men al-zµala@l, J. S®al^ba@ and K. Ayya@d, eds., 9th edition, Damascus, 1980; tr. R. J. McCarthy as Freedom and Fulfillment, Boston, 1980. Taháa@fot al-fala@sefa, tr. and ed. M. E. Marmura as The Incoherence of the Philosophers, Provo, Utah, 1997.

Studies: M. Fakhry, Islamic Occasionalism and Its Critique by Averroes, Maimonides, and Aquinas, London, 1958. R. M. Frank, Al-Ghaza@l^ and the Ash¿arite School, Durham and London, 1994. G. L. Gardet and G. C. Anawati, Introduction aà the‚ologie musulmane, Paris, 1948. D. Gimaret, La doctrine d'al-Ash¿ar^, Paris, 1990. G. F. Hourani, "A Revised Chronology of al-Ghaza@l^'s Writings,@" JAOS 104, 1984, pp. 289-302. F. Jabre, La notion de certitude selon Ghazali, Paris, 1958. M. E. Marmura, "Al-Ghaza@l^ and Demonstrative Science," Journal of the History of Philosophy 3/2, 1965, pp. 46-75. Idem, "Ghaza@l^'s Chapter on Divine Power in the Iqtisa@d," Arabic Science and Philosophy 4, 1994, pp. 270-315. Idem, "Ghaza@lian Causes and Intermediaries," JAOS 115/1 1995, pp. 89-100. K. Nakamura, "Was al-Ghaza@l^ an Ash¿arite?" Memoirs of the Research Department of the Tokyo Bunko 51, 1993, pp. 1-24. J. Obermann, Der philosophische und religiose Subjectivismus Ghaza@l^s, Vienna and Leipzig, 1921. F. A. Shehadi, Ghazali's Unique Unknowable God, Leiden, 1975. W. M. Watt, Muslim Intellectual, Edinburgh, 1963.