Al-Ghazālī's Critique of Causality as Epistemological Skepticism — Epoche C2
The passage at issue In the seventeenth of the twenty discussions of the Tahāfut al-Falāsifa (The Incoherence of the Philosophers), completed in 1095, Abū Ḥāmid al-Ghazālī (c. 1056–1111) attacks one claim: that the tie between contact with fire and the burning of cotton is a necessary tie, and that the human intellect can establish it as such by demonstration. Everything the chapter is famous for — the denial of natural efficacy, the vindication of miracle, the supposed anticipation of Hume — hangs on how that attack is read. The reading defended here is that its load-bearing move is epistemic rather than ontological: al-Ghazālī's primary target is not the existence of causal necessity but the philosophers' claim to know it, and he presses that target using the philosophers' own standard of proof. Two corrections to the way this note previously presented its material, since both bear on the argument. First, the passage usually set out here as a quotation — running from "the connection between what is habitually believed to be a cause and what is habitually believed to be an effect is not necessary" through to "it is God who creates the burning when the fire touches the cotton" — is not a single sentence of al-Ghazālī's. The opening clause renders his thesis statement; the closing one compresses a conclusion he reaches several paragraphs later, and reaches only under the first of two defences he goes on to offer. Running them together makes the chapter look like a flat assertion of occasionalism, which is exactly the reading in dispute. Second, Richard Frank's monograph Al-Ghazālī and the Ashʿarite School appeared in 1994, not 1992; the 1992 date attached to it in the earlier version of this note belongs to a different study of his. What al-Ghazālī actually opens with is a criterion, not a verdict. For any two items where the one is not the other, and where neither the affirmation of the one entails the affirmation of the other nor the negation of the one the negation of the other, the existence of the first does not necessitate the existence of the second, and the non-existence of the first does not necessitate the non-existence of the second. Fire and burning are two such items: they are distinct existences, and neither is contained in the concept of the other. So the necessity denied in the opening sentence is entailment ( luzūm ) between distinct existences — a logical relation — and not, at that stage, natural efficacy. Keeping those two apart is the whole of the interpretative work that follows. Avicenna's necessitarianism, stated in his own terms Before the criterion can be seen to bite, the position it is aimed at has to be stated accurately, and here the earlier version of this note misdescribed it. It called Avicenna's system one in which "all existence emanates necessarily from the One". "The One" is Plotinian vocabulary; Avicenna (Ibn Sīnā, c. 980–1037) characterises his first principle modally, and the modal characterisation is what carries the argument. Everything that exists is either necessary of existence in itself ( wājib al-wujūd bi-dhātihi ) or possible of existence in itself ( mumkin al-wujūd bi-dhātihi ). Exactly one item falls under the first description. Everything else is possible in itself and, once its complete cause obtains, necessary through another ( wājib bi-ghayrihi ). From this Avicenna draws the maxim that al-Ghazālī fastens on: a thing does not exist so long as it has not become necessary. Bare possibility never issues in existence; what tips a possible thing into being is the completion of its cause, and when the cause is complete the effect cannot fail. That is a thesis about the world. Avicenna joins to it a thesis about knowledge, and it is the conjunction that al-Ghazālī must break. In the Aristotelian architecture Avicenna inherited, scientific knowledge is knowledge through causes, and demonstration ( burhān ) — syllogism from premises that are certain, prior to and explanatory of the conclusion — yields certainty ( yaqīn ) about them. The sublunary world is furnished with forms bestowed by the tenth separate intellect, the Giver of Forms ( wāhib al-ṣuwar ), upon matter that celestial motion and elemental mixture have disposed to receive them; the resulting natures then act with the necessity their causes confer. It is fire's nature to burn, and the intellect that grasps the nature grasps why the cotton must blacken. The claim to be defeated is therefore conjunctive: there is a necessity, and we have demonstrative access to it. An argument that touches only the first conjunct leaves the philosophers' science intact; an argument that touches the second brings it down whatever the truth about natures turns out to be. The argument from underdetermination Al-Ghazālī's reply concedes the observations and denies that they reach the conclusion. Set out as an argument rather than as a series of assertions, it runs: The philosophers claim demonstrative knowledge that contact with fire necessitates the burning of cotton. A demonstration proceeds from premises that are either self-evident or derived from self-evident premises. "Contact with fire is not accompanied by burning" is not self-evidently false. No contradiction can be extracted from it in the way one is extracted from "this body is at once wholly black and wholly white". So the required premise is not self-evident. It must therefore rest on observation. Observation records that burning occurs at the contact. It does not record that burning occurs through the contact, nor that nothing else is its agent. At least three hypotheses fit that record exactly: (a) fire is the efficient agent; (b) fire is the occasion and God the agent, whether without intermediary or through the angels charged with such acts; (c) fire has a genuine power to burn, conferred on it by God, which God may withhold on a given occasion. Observation therefore does not select (a) from among (a), (b) and (c). The demonstrative claim