dawn at which [it might break], he would be permitted either to abstain or to eat. If he abstained at dawn, there would be no sense in describing abstention as recommended; rather, [if he were uncertain whether daybreak had occurred, he would be required to abstain].

Most of Ahmad’s recorded statements support only this position: he recommended fasting that day and would himself fast it, but did not make it obligatory. In this, he merely followed what he related from the Companions in the collections of legal questions transmitted by his son Abdullāh, al-Fadl ibn Ziyād al-Qattān, and others. He followed what he had related from Abdullāh ibn Umar and others like him. What is reported from them is that they would fast when the sky was overcast, but did not make fasting obligatory. Most people did not fast, and they did not reproach them for refraining.

The reason he did not recommend fasting when the sky was clear—and indeed prohibited it—was that both the presumption and the apparent circumstances indicated that the crescent had not appeared. Fasting that day would therefore mean beginning Ramadan one day early, which the Prophet—may Allah bless him and grant him peace—forbade.

Two reports are related from him as to whether a day of cloud cover is called a day of doubt. His followers likewise differed on this matter.

As for a day on which the sky is clear, he regarded it either as a day of doubt or as certainly belonging to Sha‘bān, and he prohibited fasting it without hesitation. The foundational principles of the Sharia provide stronger support for this position than for any other. When it is uncertain whether something is obligatory—as when one is uncertain whether zakat, an expiation, a prayer, or something else is obligatory—it is not obligatory to perform it, nor is it recommended to omit it. Rather, performing it is recommended as a precaution.