Echoes of Doubt: Al-Ghazālī's Yaqīn and the Augustinian Answer to Scepticism — Epoche C2
In the autobiographical treatise he composed near the end of his life, al-Munqidh min al-ḍalāl , the Deliverance from Error, al-Ghazālī reports that for close to two months he was unable to believe anything he said: a sceptic in fact, as he puts it, though not in speech or outward profession. That episode, and the way he says it ended, are the subject of this essay. The thesis is the one the essay has always carried — that al-Ghazālī's route out of doubt and Augustine's have a common structure, and that the structure is imposed by the regress both men ran into rather than by anything either tradition inherited from the other. What has to change is the framework in which the comparison was set, because that framework named two positions which, under the descriptions given, nobody holds. 1. A framing to be withdrawn before the comparison can be made The essay in its earlier form opened by contrasting "Islamic kalām coherence theories of truth, particularly those articulated by the Ash'arites" with "Western analogues, such as Augustinian coherence theories". No such positions are on offer in either tradition. A coherence theory of truth holds that a proposition is true in virtue of its membership in a maximally coherent set of propositions, and neither party to this comparison holds anything of the kind; both hold that truth is fixed independently of any human system of beliefs, by God. What the Ashʿarite school actually maintains is occasionalism: God is the sole efficient cause, created things possess no causal power of their own, and what appears as natural necessity is the regularity of God's own practice. Its epistemology is correspondingly foundationalist rather than coherentist. It rests on ʿilm ḍarūrī , necessary knowledge — items such as the deliverances of sound sense and the first principles of the intellect, which are created immediately in the knower and are not reached by inference from anything else. A structure that terminates in items justified by no other item is the opposite of a coherentist one. Augustine's is foundationalist in the same broad sense: truth is the immutable divine Truth, and the mind judges by a light it does not itself supply. So the contrast the essay set out to dissolve was not a contrast between two real positions, and the argument is better without it. One further term in that opening needs its plain meaning, since the essay used it without one. Al-ʿaql al-faʿʿāl , the Active Intellect, is not a general name for external epistemic authority. It is the tenth and lowest of the separate intellects in the emanative cosmology of al-Fārābī and Ibn Sīnā, the giver of forms, from which a suitably prepared human intellect receives the intelligibles. Al-Ghazālī's rejection of it, argued in the Tahāfut al-falāsifa , is the rejection of one specific Avicennan doctrine, and it does not by itself amount to a general suspicion of anything outside the mind. 2. The crisis: what al-Ghazālī doubted, and by what argument The Munqidh begins the sceptical passage by fixing a criterion, and the whole subsequent argument is driven by it: certain knowledge is that in which the thing known is disclosed so completely that no doubt attaches to it and no possibility of error accompanies it. He gives the test in a form worth reproducing exactly, because the earlier version of this essay reproduced it inexactly. A man who turned a stone into gold and a staff into a serpent, al-Ghazālī says, would not thereby give him any reason to doubt that ten is more than three. The earlier version had the example as "two is greater than one"; the point is unaffected but the text says ten and three, and the number of a first principle is the sort of thing a reader will check. Against that criterion, two demolitions follow in order. Sense. Sight judges a shadow to be at rest; watched for an hour, it is found to have moved, and to have moved continuously rather than in jumps, so the original judgement was wrong and nothing in sight itself revealed the error. Sight judges a star to be the size of a coin; geometrical demonstrations establish that it is larger than the earth. In each case the verdict of sense is overturned by the intellect, and overturned decisively. Intellect. The demolition of sense is what makes the second step available, and al-Ghazālī draws it out. Sense was confident and was convicted by a higher judge. What then secures the judge? He entertains the possibility of a further faculty, standing to the intellect as the intellect stands to sense, whose arrival would show the intellect's judgements to be as unreliable as sense now appears; and he supports the possibility with the dream: while dreaming, one takes things to be firm and real, and on waking they are shown to be nothing, so the waking state may stand to some further state as the dream stands to waking. The Sufis' report of a condition in which they see what contradicts the intellect is offered as a candidate for that further state. The step that gives the crisis its philosophical interest is the one about what could end it. Al-Ghazālī says the disease could not be treated by argument, and gives the reason: any proof would have to be composed out of the primary knowledges, and it is exactly those whose standing is in question, so no proof can be constructed that does not presuppose what it is meant to establish. This is a clean statement of the ancient problem of the criterion, and it is the reason his exit takes the form it does. The earlier version of this essay glossed the resulting condition as a "mild sickness"; the text does not support the adjective, and the description he gives — unable for nearly two months to affirm what he continued to say — is not that of a mild condition. The exit, when it comes, is expressly not an inference. God restored him, he writes, not by an ordered proof or an arranged argument but by a light which God cast into his breast, and that light, he adds, is the key to most knowledge. The pr