yet even this is highly improbable, for at that point the moon would already be larger than the normal arc of visibility. Although its size does fluctuate, it is—Allah knows best—unlikely that it could become large enough to be seen while the sun is still above the horizon. In any case, as we have said, the decisive standard here is empirical observation, and it makes no difference whether that occurs before or after noon; the only factor that counts is whether or not the sun has set.

Their disagreement over how certainty of sighting is established stems from two possible avenues: sensory perception and transmitted report.

1. The avenue of sensory perception
   The scholars unanimously agree that anyone who personally sees the crescent of Ramadan must fast. ʿAṭāʾ ibn Abī Rabāḥ alone held that one does not fast unless someone else sees it along with him. They also differed over whether a person may break the fast on the strength of his own solitary sighting. Mālik, Abū Ḥanīfah, and Aḥmad said that he may not. Al-Shāfiʿī, followed by Abū Thawr, said that he may. This latter view, however, makes little sense, for the Prophet—peace and blessings be upon him—made both fasting and breaking the fast contingent on sighting, and sighting is by definition a sensory act. Were it not for the consensus allowing fasting on the basis of a report of sighting, the ḥadīth’s plain wording would make it difficult to oblige fasting on the basis of mere reportage. Those who distinguished between the crescent of beginning the fast and that of ending it did so only as a preventive measure, lest the impious claim to have seen the crescent and break the fast without actually having seen it.
   For this reason al-Shāfiʿī said: “If he fears suspicion, he should abstain from food and drink outwardly while inwardly holding that the fast is over.” Mālik, in an unusual stance, said that anyone who breaks the fast after seeing the crescent alone must both make up the day and offer expiation, whereas Abū Ḥanīfah required only that he make it up.

2. The avenue of transmitted report
   Here they differed over both the number and the qualifications of those whose report of the sighting must be accepted. Mālik held that it is not permissible to begin or end the fast with fewer than the testimony of two upright men. Al-Shāfiʿī, in the narration of al-Muzanī, ruled that the month may be commenced on the testimony of a single man but may not be concluded with fewer than two. Abū Ḥanīfah said that if the sky is overcast the testimony of one is accepted, but if it is clear over a large city nothing short of a sizable multitude will suffice; another report from him accepts the testimony of two upright men when the sky is clear. Mālik is also reported to have accepted the testimony of two only when the sky is cloudy.
   They are unanimously agreed that for ending the fast (the crescent of Shawwāl) at least two witnesses are required—except Abū Thawr, who, unlike al-Shāfiʿī, made no distinction between the start and the end of Ramadan.

The root of their disagreement lies in the varying narrations on this issue and the question of whether such reports are to be treated as formal testimony, which requires a set number of witnesses, or as solitary ḥadīths, which do not.

As for the narrations themselves, one is the report recorded by Abū Dāwūd from ʿAbd al-Raḥmān ibn Zayd ibn al-Khaṭṭāb, who addressed the people on a day of doubt, saying, “I have sat with the Companions of the Messenger of Allah—may Allah bless him and grant him peace—and asked them, and all of them told me that the Messenger of Allah—may Allah bless him and grant him peace—”