—together with the text itself—al-Bukhārī would reply, “I do not recognize it,” and he continued to say so until they had finished the entire hundred. The laymen in the gathering began to mutter, “This man knows nothing! A hundred ḥadīths are put before him and he keeps saying, ‘I do not recognize it,’” meaning: I do not recognize this chain for this particular text. Then al-Bukhārī, may Allah have mercy on him, rose and recited every ḥadīth with its correct chain, one after another, until he completed all one hundred. At that point they realized he was a marvel granted by Allah in the art of memorization, and they all admitted his mastery and deferred to him.

This is what we call “inverting the chain of a text,” that is, grafting the chain of one report onto the text of another. Most of the time it is done only as a test, though it can also be used deceitfully: a man may wish to circulate a ḥadīth whose own chain is worthless—all of its transmitters being weak, for example—so he takes the chain of a sound ḥadīth and fastens it to that text. This is a form of tadlīs (concealment), but by a different method.

There is another category, namely inversion of the text itself, and this is what concerns the jurists, whereas inversion of the chain concerns the ḥadīth critics, for they examine whether the chain is sound and whether a ḥadīth can be authenticated through it. The jurists, however, focus on inversion of the text because it alters the legal ruling; their concern is with the import and evidence of the wording.

Textual inversion occurs when some narrators have the wording of a ḥadīth turned around on them, so they transmit it in an incorrect form. An example is the ḥadīth of Abū Hurayrah, may Allah be pleased with him, in which the Prophet, peace and blessings be upon him, said: “Seven will be shaded by Allah in His shade on the day when there is no shade but His…” 1 The ḥadīth continues to list them, and within it