so, for them, the presence of the Messenger—may Allah bless him and grant him peace—became equivalent to his absence in questions concerning Allah and knowledge of His lordship. Indeed, according to their position, his presence is more harmful than his absence, since they have learned nothing from him and have had to dismiss what he brought, whether by denying it, consigning knowledge of its meaning to Allah, or reinterpreting it. This has been discussed at length elsewhere.

If they say:

“It is inconceivable that one could know him to have reported anything contrary to reason, for he is above such a thing, and it is impossible for him.”

They are told:

“This is an acknowledgment on your part that rational evidence cannot conflict with evidence from revelation.”

If they say:

“We meant only a conflict with something thought to be evidence that is not evidence at all, or with evidence that yields only probability because some of its premises admit uncertainty, whether in its chain of transmission or in the text of the report—for example, the possibility that the reporter has lied or erred, or that the wording admits two or more meanings.”

The reply is as follows:

If you define evidence from revelation as something that is not actually evidence, so that believing it to establish a conclusion is ignorance, or as something thought to be evidence that is not evidence, then rational evidence that conflicts with revelation can likewise be defined as something that is not actually evidence, so that believing it to establish a conclusion is ignorance, or as something thought to be evidence that is not evidence.

In that case, something of this kind—even if its proponents call it a rational demonstration or a conclusive rational proof, when it is not actually evidence or establishes its conclusion only with probability—if it conflicts with what is evidence