In lieu of an abstract, here is a brief excerpt of the content:

Journal of the History of Philosophy 38.4 (2000) 479-502

[Access article in PDF]

Possible Worlds in the Tahâfut al-Falâsifa Al-Ghazâlî on Creation and Contingency

University of Helsinki


This article is the second half in an inquiry into the debate between al-Ghazâlî (1058-1111) and Averroes (1126-1198) on the metaphysical basis of modalities. The first article focused on Averroes' exposition of the Arabic Aristotelian position on the eternity of the world and the modalities, as found in the Tahâfut al-tahâfut, the "Incoherence of the Incoherence [of the Philosophers]" from ca. 1180.1 In this paper I propose to sketch the outlines to a very different conception suggested by al-Ghazâlî in the Tahâfut al-falâsifa (1095) and quoted verbatim and commented upon by Averroes in the Tahâfut al-tahâfut.2 In contrast to the naturalistic model of modalities propounded by [End Page 479] Averroes, al-Ghazâlî develops an interpretation of possibility in which conceivability and co-assertability are central. In several key respects al-Ghazâlî's approach mirrors that of John Duns Scotus in the fourteenth century: it makes possible, for instance, a distinction between conceptual and natural possibilities in a manner reminiscent of later Latin medieval theory, and, reflecting this, a distinction between the (non-temporal) moments of conception and causation in the divine mind. The parallels between al-Ghazâlî's and Augustine's basic conceptions, meanwhile, highlight the theological motives lying at the heart of the new thinking.

As in the previous article, the dialectical form of Averroes' later Tahâfut al-tahâfut is used for critical insight. Averroes' criticism of al-Ghazâlî's modal theoretical innovations gives rise to an important question about the lengths to which al-Ghazâlî's proposals could be taken. The issue is to an extent unresolveable; one would do well to note why. Whereas Averroes' formulations in the Tahâfut al-tahâfut are made readily understandable against the backdrop of his Aristotelian commentaries and wider systematic teaching, it is less easy to determine how far al-Ghazâlî is willing to develop his own ideas. The materials given in the Tahâfut al-falâsifa leave the impression of being inconclusive and open-ended—perhaps deliberately so—and the ambiguities are mirrored in al-Ghazâlî's other texts.3

Taken as they stand, al-Ghazâlî's formulations retain a substantial Avicennian residue, for better or for worse. This makes for a curious mixture of old and new in his overall scheme, a feature which Averroes can exploit by criticizing al-Ghazâlî from a strict Aristotelian viewpoint (just as he criticized Avicenna), while all the time maintaining that his adversary's proposals do not add up to a coherent and full-bodied alternative to that venerable tradition. But Averroes' criticisms do not touch on the way al-Ghazâlî's suggestions, vague as they may be, do represent a genuine breakthrough. Al-Ghazâlî's interpretation of the modal terms bypasses the connection traditionally made between true possibility and temporal instantiation. This allows for a freer handling of counterfactual possibilities, an important feature in any refiguring of the limits of possibility.


Al-Ghazâlî's innovations have their root in the idea of God freely choosing (arbitrating) between alternatives equal to him. This simple kernel of an idea, seated in the kalâm conception of possibility, allows for a recognition of the genuineness of synchronic alternatives: a seemingly small shift, but all-important [End Page 480] to the ensuing debate. Al-Ghazâlî has defended the notion of God's genuine choice earlier,4 but in refuting "the third proof of the philosophers for the world's eternity," he begins to make systematic use of the notion of possibility. A brief recapitulation of the philosophers' argument will serve to recall the reasoning from which al-Ghazâlî takes his cue. The philosophers, in al-Ghazâlî's words, had...