Al-Ash’ari doesn’t reject the use of reason, but he does wish to reject the Mu’tazilite approach, but not to stop doing theology. Rather, he wished to use it for the defense of a more traditional doctrine. He uses reason conspicuously. But the relation between reason and revelation is approximately the opposite way round from how al-Jabbar describes it. Al-Ash’ari operates on the assumption that the Qur’an and the Traditions are to be interpreted literally wherever this is possible. He acknowledges that the Qur’an does also sometimes speak metaphorically, but he thinks the literal interpretation should be used when it isn’t impossible.
His second major criticism of the Mu’tazilites is that they hold the Qur’an to be created, whereas al-Ash’ari holds that it was recorded in time, but is itself eternal. The most important point for our purposes is that al-Ash’ari does not think we are justified in holding revelation to some standard of interpretation external to it. God gives guidance, he says, to the faithful, and not to the unfaithful (infidels). The Qur’an has a verse that teaches that the Prophet warns both the one who follows and the one who does not, the unfaithful one. Al-Ash’ari concludes from this that guidance and warning are different. The revelation warns both faithful and unfaithful, but only guides the faithful, and there are some warnings also that are given only to the faithful. The point is just that al-Ash’ari can’t allow what the Mu’tazilites assert, namely, that the guidance gives all human beings what they then recognize as means to what their reason already prescribed for them.
We can relate al-Ash’ari’s position about the sources of theological knowledge to the four traditional sources of Islamic law: the Qur’an, the Traditions, the consensus of the faithful, and analogical deduction from Scripture. Of these the first two are given by revelation. For al-Ash’ari, the third (consensus), as it applies to theological knowledge, is also given by special divine grace. But this is not because of a general truth about communities of religious believers, but because of a special dispensation given to Muslims. The fourth source, analogy, is likewise strictly restricted in its theological use to what is implied by the revealed texts themselves. Sometimes we can tell from a scriptural prescription what God’s reason is for prescribing in this way, and sometimes we can apply that reason to cases analogous to the original case. But the point is that al-Ash’ari, in accepting these four traditional sources, is not putting them under two mutually independent headings, revelation encompassing the first two and reason the second two. Rather, the second two depend for their authority on special revelation.