Time told by the moon

Find in each source what sets the start of a month or a year: something seen in the sky, or something remembered from the past. Note which of the festivals and rites you have met this term depend on that start.

Ibn al-Athir's al-Kamil says that in the year 16 Umar ibn al-Khattab (may God be pleased with him) set the dating of the calendar on the advice of Ali ibn Abi Talib (may God be pleased with him), and that the start of the Islamic calendar was made the Prophet ﷺ's migration from Mecca to Medina ². In the chapter on dating in Sahih al-Bukhari, Sahl ibn Sa'd says that they did not count from the Prophet's mission or from his death, but only from his arrival in Medina ³. The Encyclopaedia of Islam adds that the era does not begin on the day Muhammad reached Medina, but on the first day of the lunar year in which this happened, reckoned to coincide with 16 July 622 ⁴. So the three sources answer the question "counted from where?" with a remembered event, not with something seen in the sky.

How is a month established by the crescent? The Kuwaiti encyclopaedia of jurisprudence says Ramadan is established by sighting its crescent, or by completing Sha'ban to thirty days when sighting fails. It says the jurists differ on the number of witnesses: the Hanafis, Shafi'is and Hanbalis accept one just witness, the Malikis require two. It adds that the jurists agree the end of Ramadan, the crescent of Shawwal, needs two witnesses ^5 6 7 8. The Encyclopaedia of Islam, on the Shafi'i account, says the general obligation begins on 1 Ramadan after thirty days of Sha'ban, or after twenty-nine if the judge has accepted the testimony of one trustworthy person that he saw the new moon ⁹. The museum prefers no school.

It is easy to suppose year 1 began on the day the Prophet reached Medina. The Encyclopaedia of Islam says the era does not begin on that day but on the first day of the lunar year in which it happened, and that this result assumes that no intercalary months occurred after the Hijra, which is by no means certain ⁴. In its article on the Farewell Pilgrimage it names the abolition of intercalation and a purely lunar year among the arrangements made on that occasion, and says that because the lunar calendar has no intercalary month, the pilgrimage falls ten or eleven days earlier each year and runs through the whole cycle of the seasons in thirty-two to thirty-three years ^10 11.

The report of the dating in the year 16 comes from Ibn al-Athir's al-Kamil, a history ². The record of the Hijri calendar says the reports differ over whether the decision followed a consultation with the Companions and over which year it was, and that this needs a second source before it can be treated as established. The definition of the calendar as a lunar one counted from the Hijra comes from al-Suyuti's Tarikh al-Khulafa ¹². The record of the Hijra says historians debate the exact sequence of events, so its dates are marked as traditional ¹³. The conversion of dates from Hijri to Gregorian is, in the museum's words, a derived calculation, not taken from a source.

These sources give two starting points: a sighting that begins each month, and a remembered event that begins the count of years. Which of the two would a community need to agree on more often, and what in the sources shows how disagreement about each is settled?

• The crescent: its definition and the times the crescents were made, in the Kuwaiti encyclopaedia.
• The Hijri calendar: what it fixes, and the disagreement over choosing the Hijra as its start.
• Umar's decree: the report for the year 16 in Ibn al-Athir.
• The Hijra: the accounts, their differing dates and the encyclopaedia's article.
• Ramadan: how the month is established and how many witnesses the schools accept.