History and legend in the origins of Islam (II)

Part of the Arabic inscription in the Kufic script inside the Dome of the Rock | Source: Lost Islamic History
Part of the Arabic inscription in the Kufic script inside the Dome of the Rock | Source: Lost Islamic History

In the first article of this series, we saw what Middle-East Christian texts from the 7th century told about the Arabian invaders that abruptly took those wealthy territories that during the previous centuries had been disputed between the Romans and the Persians (first, Palestina, Syria and Mesopotamia; a little later Egypt and Persia itself). In a nutshell, they identified the invader forces as guided by a ‘prophet’, though during the first decades of the invasion Christians referred to the Arabs’ faith as something vaguely close to Judaism, and only one hundred years after the supposed death of Muhammad there is some reference in those texts to the content of something similar to the Qur’an. Perhaps we might think that contemporary Arabic sources would be more explicit about the religious beliefs of those people, but this would prove to be a vain expectation, as we shall immediately see. Let’s briefly describe the most important pieces of evidence. The data are taken mostly from Spencer (2012) 1 and Holland (2012) 2.

1) Probably the oldest preserved reference that uses the Islamic calendar is a receipt of the exaction of sixty-five sheep from the Greco-Egyptian city of Herakleopolis, which is written both in Greek and in Arabic, and with the dates in the two corresponding calendars (22nd AH, 632 CE). According to the tradition, Arabs had started to count the years from the Hijra (the escape of Muhammad and his followers from Mecca to Medina in 610 CE) only five years before the date recorded in that document, in 627/28. Unfortunately, nothing on it refers to what the counting might mean, but it is clear that, being a date so close to the conventional starting year, the memory of the ‘inaugurating’ event wouldn’t have been lost or severely distorted, as compared, e.g., to the case of the Christian calendar, which started to be used more than five centuries after the birth of Jesus. Hence, this is a reason supporting the historicity of the Hijra, though of course not applicable to all of its traditional details.

2) Another almost contemporary text, recently discovered, is an inscription on stone, in the desert to the south of Palestine, mentioning that it was written: “at the time Umar died, in the year 24”. Umar was the second Caliph, and the traditional date of his assassination coincides with that of the inscription.

3) From about fifteen years later it dates another inscription, this time in a bath-house in the Syrian city of Gadara (close to the Lake of Galilee). Although the inscription is in Greek, it identifies Muawiya (the first Umayyad caliph) as “servant of God and leader of the protectors”, and gives the date “42 following the Arabs”. The inscription lacks any clear reference to Muslim religion, refers to the Arab rulers as ‘protectors’ (supposedly) of the Christian Byzantine community to which the baths belonged, and gives no religious interpretation of the year’s number; all this allows to cast some sceptical doubts about whether those Arabs were practicing something like orthodox Islam. But most surprising are the facts that the inscription does not contain any reference at all to Muhammad, and is preceded by a cross.

4) Another inscription, this time in Arabic, found in Karbala (Iraq), dates from 64 AH (683 CE), and, though in this case it starts by the traditional Muslim invocation “In the name of Allah the Merciful, the Compassionate”, later it refers to Allah as “the Lord of Gabriel, Michael and Asrafil” (three Biblical angels), again without any mention to Muhammad. Another curious fact about it is that it refers to three moments of praying (as the Qur’an itself states), not to the five that became the rule in the following century. An almost contemporary inscription found in Ta’if, near Mecca, makes also no reference at all to Muhammad. This scheme is all-pervasive in almost all official inscriptions in the Arab world dating from the first century of the Islamic era.

5) Equally surprising is the fact that the coins minted by the first caliphs, rarely include the word “Muhammad” (and when they do it, it isn’t clear whether it functions as the Prophet’s proper name, or in the sense of “let the Caliph be praised in the name of God”, because that “Muhammad” means “blessed” or “praised” in Arabic), but, contrarily to Islamic law, they represent a human figure (either the Caliph or –possibly– the Prophet)… which is carrying a cross! One coin from Muawiya’s time includes the same figure but with (for the first time) a tiny crescent topping the cross.

6) Probably, the first Islamic text beside the Qur’an is the inscription in the mosaics of Jerusalem’s Dome of the Rock, completed by 691 CE. It includes fragments of the Qur’an (though not in a completely literal way), and explicitly mentions Muhammad, but curiously most of its text is devoted to criticizing the Christian claim that Jesus is the son of God, and actually the portion of the text that is devoted to Jesus (and to praise his role as a messenger of God) is much bigger than the one devoted to the own Islam’s Prophet, of whom it only expresses the laconic claim that “Muhammad is the messenger of God”, which (as in the case of the coins) some scholars interpret not as a reference to the Arabian Prophet, but as a generic blessing to (whoever happens to be) the messenger of God, perhaps Jesus himself, or some other Biblical or post-Biblical prophet, or, more probably, to all of them.

After having devoted the first article of this series to present what Mid-Eastern Christian chroniclers contemporary of the Arab invasions wrote, it would be natural to expect that this second instalment would include the same about Arab chroniclers. The sad fact is those chronicles simply didn’t exist, at least as written works. The first Muslim historians wrote more than one century after Muhammad’s life, and even the books of those have not directly arrived to us, but only in quotes and mentions made by authors from at least many decades well into the second century of the Islamic Era. The Qur’an itself is the only Arabic literary source dating from the 7th century CE, and we will examine some of the philological research about it in our next article.

Hence, the truth is that all knowledge about the first century of Islam from ‘Islamic’ sources depends on the oral tradition. The first Arab historians made an art of the certification of the isnads, the chains of transmitters that supposedly were passing the hadith (a saying or a deed of the Prophet or of someone related to him), but obviously there were no absolutely reliable means by then of carrying out that certification without error, and, furthermore, occasion for corrupting or even inventing hadiths and isnads alike would abound in a time (more than one century) when the Islamic orthodoxy, institutions and nuclei of power were just taking shape. It is told that Bukhari, one of the first hadith collectors in the 9th century CE compiled about 300.000 such stories, of which he chose to publish only about 7.500 as valuable, but with only around one third that he himself considered “authentic”; so, less than one out of thousand.

There is one particularly compelling argument (Jansen, 2008) 3 that serves to shed doubt on the credibility of most of the hadiths. It is the fact that in the first biography (Sira) of Muhammad, written by Ibn Ishaq in mid 8th century CE, dates are meticulously recorded. This fact apparently gave a high credibility to Ibn Ishaq’s narration, both under the eyes of subsequent Islamic historiographers and from the point of view of Western scholars. But there is a deep problem with Ibn Ishaq’s dates. The Islamic lunar calendar substituted a pre-Islamic Arabian, also lunar one, by 629 CE (still in Muhammad’s life), the main difference between both is that the pre-Islamic calendar included a ‘leap month’ every three years to keep the pace of the solar calendar (for twelve lunar months add only up to 354 days). Muhammad forbade the practice of including leap months, and hence the Islamic official calendar has a year which is about 11 days shorter than the solar year, which means that 34 Islamic years are equivalent to about 33 –Julian or Gregorian–solar years. But curiously, not a single event previous to 629 CE narrated by Ibn Ishaq takes place during a leap month! According to Jansen, the most sensible explanation of this is that the hadiths containing those events were invented in a time when people had just forgotten that leap months had existed at all.


  1. Spencer, R., (2012), Did Muhammad Exist? An Inquiry into Islam’s Obscure Origins, Washington, ISI Books.
  2. Holland, Th., (2012), In the Shadow of the Sword. The Battle for Global Empire and the End of the Ancient World, Little, Brown.
  3. Jansen, H., Mohammed: Eine Biographie, München, Beck.

Written by

Leave a Reply

Your email address will not be published.Required fields are marked *