Journal of the History of Philosophy

Ghazali and Demonstrative Science MICHAEL E. MARMURA I MEDIEVALISLA_MICtheologians subjected Aristotle's theory of the essential efficient cause to severe criticism and rejected it. This criticism and rejection finds its most forceful expression in the writings of Ghazali (al-Ghaz~li) (d. 1111).1 In his Tahafut al-Falasifa (The Incoherence of the Philosophers), he argues on logical and empirical grounds that the alleged necessary connection between what is habitually regarded as the natural efficient cause and its effect cannot be proven. He does not, however, take an agnostic stand on this question but proceeds to affirm two points: (1) that there is no causal agency in natural things; (2) that all natural events are the direct creation of God. Ghazali, moreover, does not confine such divine action to the realm of the inanimate and the irrational, in his al-Iqtis.~d fi-l-Ftiqad (The Golden Mean of Belief), he affirms the doctrine that the individual human act, like any other occurrence in the world, is also the direct creation of God. Causal efficacy resides in God alone. But the divine act, for Ghazali, is not an essential act. It does not proceed as the necessary consequence of divine essence. It is the arbitrary decree of the divine will, an attribute coeternal, but not identical with divine essence. Ghazali thus denies essential action altogether. In all this, he gives expression to the occasionalism of the Islamic school of dogmatic theology (kal~m) to which he belonged, the Ashcarite school. Ghazali had also a keen interest in logic. He wrote logical treatises for his fellow theologians, encouraging them to master this art as a tool to rebutt their doctrinal opponents. The~ treatises, in the main, sum up and explain the logic used by the Islamic philosophers--a logic that reflects the Stoic and Aristotelian traditions. Ghazali recognizes formal logic to be philosophically neutral, and hence in no way conflicting with his theology." He does not confine himself to formal logic in these expositions, however, but discusses Aristotelian demonstration. And it is in his statements on demonstration that we seem to be confronted with paradox. Demonstrative science, as understood by the Islamic philosophers, rests on the theory of essential causes and functions in terms of it. Indeed, in his reply to Ghazali's rejection of essential efficient causes in nature, Averroes (Ibn Rushd) (d. 1198) argues Texts frequently referred to in the notes will be abbreviated as follows: Demonstration: Ibn Sin~, Al-Shifa~; Logic V.; Demonstration, ed. A. E. Affifi,revised by I. Madkur (Cairo, 1956). Iqti~dd: A1-Ghaz~li, Al-Iqtisad Fi-l-ICtiqad(Cairo: no date). Metaphysics: Ibn Sin~, A1-Shif~: al-Ilah~yat (Metaphysics), ed. C. C. Anawati, S. Dunya, and S. Zayd, revised by I. Madkur (2 vols.; Cairo, 1960). MI: AI-Ghaz~lI,MiCyaral-CIlm,ed. S. Dunya (Cairo, 1961). TF: A1-Ghaz~ll, Tahafut al-Falasifa, ed. M. Bouyges (Beirut, 1927). TT: Ibn Rushd (Averroes), Tahafut al-Tahafut, ed. M. Bouyges, Beirut, 1930). 2 TF., pp. 15-17. [1831 184 HISTORY OF PHILOSOPHY that such a rejection would render demonstrative science impossible. We would have no true knowledge; at best, only opinion. 3For this staunch defender of Aristotle , the theory of natural efficient causality constitutes a necessary condition for demonstrative science. Yet Ghazali in his writings does not deny the claim that demonstration gives us certain knowledge about the natural order. On the contrary, he seems to be affirming this in both his Tahdfutand the logical treatise he appended to this work, Mi~jdral-CIlm(The Standardfor Knowledge).If Ghazali does in fact uphold this claim for demonstration, then he must either deny that the theory of essential efficient causality is a necessary condition for demonstrative science, or fall into contradiction. At first glance, Ghazali's position in the Micy~r,where he gives his most comprehensive treatment of demonstration, is highly ambiguous and suggests contradiction . He seems for the most part to be reproducing the essentials of the Aristotelian demonstrative theory. Although in one section he reaffirms his rejection of necessary causal connection in nature, he nonetheless continues to use such terms as "cause," "effect," "necessity" and "certainty" in what appear to be...