—at times by appealing to self-evidence, and at others by appealing to inference. As for the former, they sometimes claim that rational people attain this self-evident knowledge through testing, without any need for illustrative examples; at other times, they establish through inference that it is known self-evidently.” He elaborated their arguments on this, then said:

“The answer is to say: You claim self-evident knowledge either that perception occurs when these things are present, or that it does not occur when they are absent. The first is undisputed; the second is precisely where the dispute lies. If you allege that our denial is obstinate, we swear the most solemn oaths that, upon examining our own minds, we found our knowledge of this no greater than our knowledge that customary events will continue in their regular course—which you agree with us may be altered.”

This answer of his implies that he conceded that there is self-evident knowledge that vision occurs when these conditions are present. Yet this is where the discussion began: he reported that the Mu‘tazilah maintain that whenever these conditions are all met, perception must occur, and then said: “In our view, however, this is not necessary.” Thus, if he concedes that self-evident knowledge of perception occurring when these things are present is undisputed, he has conceded the very point concerning which he had reported