We’ve reached a point in our overview of Islamic history where we need to stop talking about history and talk historiography, or the study of how history gets studied. Because the material regarding the life of Muhammad is so slim and presents so many challenges to the historian, I feel like I would be doing readers a disservice if I were to just tell the story and move on as though there were no caveats. Things may get a little dry (drier?) here, so consider yourself warned.
So after spending two segments on the life of Muhammad, now we talk about why anything and everything we think we know about Muhammad may be wrong. Not “is,” and in fact most scholars nowadays come down on the side of generally accepting much of the narrative of Muhammad’s life as accurate, or at least close to accurate, but “may be.” The problem is simply one of sources. Historians deal mostly in textual sources: inscriptions, archival documents, narrative histories, law codes and guidebooks, etc., but they like for there to be some corroborating archaeology underlying the texts, and the fact of the matter is that for Muhammad we have nothing of the sort. There are no coins depicting Muhammad that were minted in Medina; the Arabs didn’t mint their own coins until the end of the seventh century. Archaeology at Mecca and Medina, to the extent that any has been done (and between the restrictions on access to those cities and Saudi Arabia’s ongoing wanton destruction of historical sites there, there hasn’t been enough), can’t produce anything that can be directly tied to Muhammad. No documents have been found apart from the Constitution of Medina, and that only exists as a copy of a copy; the constitution was supposedly copied by an early biographer of Muhammad, but what we have today is a later edited form of that biography. Bluntly, if you’re looking for concrete, contemporary evidence of Muhammad’s life and career, it’s really not there. It seems to me that there are four main sources for the study of Muhammad’s life, and they’re all complicated and complicating: three are internal to the Islamic tradition (the Qur’an, the Hadith, and prophetic biographies), and one is external (contemporary accounts from outside Arabia). After the jump we’ll look at each in turn.
I’ll be doing a whole entry on the Qur’an in a bit, so we don’t have to spend too much time on it right now, but it is the principle source for Muhammad in that it is the only document or item to which we can tie him, however tenuously. The biggest difficulty with using the Qur’an as a source for Muhammad is the same reason I’m not doing an entry on the Qur’an right now, which is that it can’t be attested as a written document until the end of the seventh century, and then only as inscriptions on the walls of the Dome of the Rock in Jerusalem. Muhammad recited the text orally, but there is no suggestion in any source that he wrote it down himself. Indeed, Islamic tradition contends that he was illiterate, which serves to heighten the “miraculous” origins of the beautifully literate Qur’anic text, though his previous success as a merchant (enough, recall, to get the wealthy widow Khadijah to hire him to manage her business affairs) seems to me to demonstrate that he was literate on some level. Tradition relates that others began transcribing Muhammad’s preaching and that leading figures in the community, like Abu Bakr, collected their own codices of his recitations, but if any of these ever existed they are lost to us today. Under the reign of the caliph ‘Uthman (more on him soon, but his reign lasted 644-656) it’s said that a program was undertaken to collect every transcription that could be found, with an eye toward creating a standard text, but we simply don’t see a complete Qur’an before the 8th century.
The other problem with the Qur’an as a historical source is that the Qur’an doesn’t talk about historical episodes in Muhammad’s career in any concrete way (by contrast, it relates Biblical stories, like those of Moses, Joseph, and Jesus, in some detail). What it does seem to do, however, is to respond tonally and in terms of content to the changing circumstances of Muhammad’s life, if one assumes that the traditional biography of Muhammad is accurate at least in broad strokes. So when Muhammad begins his ministry in Mecca the verses reflect a holy man preaching about divine matters to an audience that needs to be shaken up to hear the message, but once he’s leading a political community at Medina his preaching becomes much more legalistic. However, this method of categorizing Qur’anic verses according to what was happening in Muhammad’s life is also very problematic, since the written Qur’an is not organized chronologically. But more on that later. The upshot is that the Qur’an is the best source we have for the life of Muhammad–it may be chronologically close to his lifetime and it’s a text that theoretically wouldn’t have existed had he not revealed it to his followers–but it is far from being comprehensive.
Apart from the Qur’an, the most important text for Islam is the Hadith (the word means “tradition,” or “report”), the collected sayings and deeds of Muhammad when he wasn’t directly preaching the revelation he’d received from God. You could compare the Hadith to parts of the New Testament (the epistles, maybe) or Christian apocrypha (the Gospel of Thomas, say) or to the Talmud, but I would limit the comparison to its status as a sort of adjunct to the holy text (Torah-Talmud, Gospels-Epistles, Qur’an-Hadith), because comparing these things on content is really tricky. Jews and Christians don’t make the same claims to a divine origin for the Torah or the Gospels that Muslims make about the Qur’an, for starters, but I digress. The Hadith is probably the single most problematic source in the entire corpus of Islamic history, from a historian’s perspective. It purports to be a record of Muhammad’s sayings and deeds, transmitted orally (though also written down from time to time) by his followers to the generations that came after him, but nobody managed to produce a standard corpus of transmissions until the 9th century, a full 2 centuries after Muhammad died. In that time we know that vast numbers of Hadith were simply forged as they were needed, for example to justify acts of the political authorities, or to justify rebellion against those authorities, or to support or refute the claims of this or that legal/philosophical/theological movement.
Early Islamic scholars attempted to solve the problem of forged Hadith by creating a formal “Hadith science” that involved studying chains of transmission (each Hadith report is prefaced by a list of names who had transmitted that report, ideally going all the way back to someone close to Muhammad) to identify obvious forgeries or broken chains. Reports whose chains were suspect were discounted as likely forgeries (this project is considered one of the ancestors of modern critical historiography). The obvious problem is that even once a body of “reliable,” or “sound,” Hadith collections had been assembled (there are six “authentic” Hadith collections in the Sunni tradition), there’s still no guarantee that any of the reports in those collections are actually authentic. All we really know is that their chains of transmission couldn’t be easily picked apart, but a chain of transmission can be forged just as easily as the saying itself. The Hadith, unfortunately, is much more reliable as a guide to the disputes of the developing Islamic community, disputes that then inspired the creation of Hadith in support or refutation of argument, than it is as a source for Muhammad and his life.
Biographies of Muhammad
Muhammad, as the Messenger of God and the founder/leader of the Community (ummah), inspired a number of stabs at his biography, collectively known as sirah literature. It is largely from the sirah that the “traditional” account of Muhammad’s life we covered in the last two entries is derived. Biographers relied on fragmentary oral and written reports of Muhammad’s life, i.e., Hadith, to write narratives that were intended to hold Muhammad up as the ideal man and the model for a holy life.The sirah is a wonderful body of material for the historian–compiled, well-written, detailed narrative biography writing, and while they are obviously intended to support the idea of Muhammad as Messenger of God, critical reading can get beyond the places where these accounts obviously delve into myth-making or storytelling. The problem, say it with me now, is that these biographies are written too late to be reliable sources. The entire sirah genre wasn’t even conceived until 100-150 years after Muhammad’s death. If the Gospels aren’t considered reliable sources for the historicity of Jesus, and I would agree that they’re not, consider that the earliest-written Gospel, Mark, is thought to be written about 50 years after Jesus’ death, at least a half-century quicker than the earliest sirah works.
Worse, we don’t even have these earliest sirah writings; instead, what we have are edited versions of them that were done another 50-100 years after the originals. The earliest widely-attested biography of Muhammad was written by Ibn Ishaq (d. 761 or 767, and pronounced “Is-haq”), but survives only as a recension by a later writer named Ibn Hisham (d. 833), and in quotation by a handful of other writers. This is a mess, from a historian’s perspective. We’re again left with a body of material that is probably more useful as a window into the debates and discussions that were affecting the community in the mid-8th or early-9th centuries than it is as a source for Muhammad’s life. But, and this can be applied to the Hadith as well, historians are effectively forced by the lack of any other material to start with these accounts, however flawed they may be, as baselines and then try to triangulate the full story, using the Qur’an and non-Arab accounts as well as a very critical eye to piece things together. Throwing these later accounts out in their entirety can take you to some very strange places, as we’ll see in the next entry.
At last we come to a category of material for which dating is not the problem. There are several very fragmentary accounts in non-Arab/non-Islamic sources that mention, or seem to mention, the rise of a religious movement (and then empire) out of Arabia, and some reference a leader who seems to be Muhammad. These are seventh-century sources, contemporary with Muhammad and his immediate successors. For example:
- A Syriac fragment from the 630s refers to the “Arabs of Muhammad” defeating Byzantine armies
- A fragment manuscript attributed to an Iraqi Christian named Thomas the Presbyter also mentions the “Arabs of Muhammad” in reference to a victory over the Byzantines
- An Armenian bishop named Sebeos, writing in the 660s and seemingly based on his own eyewitness to events, writes about a descendant of “Ismael” named “Mahmet” who “taught [the “sons of Ismael”] to recognize the God of Abraham”
- A Greek text called the “Teaching of Jacob” talks about a “false” prophet (“prophets do not come armed with a sword”) leading the Arabs in some sort of Jewish messianic movement
The major issue with sources like this is that they must naturally lack any kind of detail because the writers couldn’t possibly have had intimate knowledge of what was happening. The “Teaching of Jacob” offers a tantalizing twist on the traditional story, that the Arabs were participating in a Jewish messianic movement, and that whoever their “prophet” was, he was still alive in 634, two years after Muhammad died, but to assume that its author was operating on actual knowledge of the Arab armies rather than hearsay and reports is probably a bad idea, and giving it too much weight can take you to those strange places I mentioned above.
When you consider the defects in the sources for Muhammad and the early Islamic community, the solution generally used in the scholarly community today is to assume that, for example, the biographic literature captures the broad strokes of Muhammad’s life, unless something can be found via other sources (non-Islamic, archaeological, etc.) that directly contradicts the biographical accounts. This is not done uncritically, by any means, and historians just avoid the supernatural elements of these biographies as outside their field of study, but it is still heavily and unfortunately reliant on some very flawed sources.
Next time: two alternate takes on Muhammad and the early community
Fred Donner’s Narratives of Islamic Origins is a crucial work that touches on aspects of this problem with the sources, although it goes beyond this period and into the origins of Islamic history-writing. Similarly Albrecht Noth’s The Early Arabic Historical Tradition: A Source-Critical Study. Jonathan E. Brockopp edited The Cambridge Companion to Muhammad, and it is very good as a survey of the field.
If primary sources are your interest, Alfred Guillaume claims to have largely reconstructed the original text of Ibn Ishaq’s biography of Muhammad and translated it into English as The Life of Muhammad. Guillaume apparently relied on Ibn Hisham but tried to reconstruct the parts that Ibn Hisham edited or removed by piecing bits of the original from quotations by other early Islamic scholars. I can’t speak to it but it sounds really interesting, if you understand that even if he did manage to reconstruct Ibn Ishaq’s original text, it would still not be a terribly reliable source for what really happened during Muhammad’s life. Several translation of Hadith exist of varying quality.
Robert Hoyland, whose Arabia and the Arabs we encountered a few entries ago, produced the most comprehensive work on non-Islamic sources on Muhammad, called Seeing Islam As Others Saw It: A Survey and Evaluation of Christian, Jewish and Zoroastrian Writings on Early Islam. Very academic but highly recommended if this interests you.
I’m intentionally skipping reading on the Qur’an because that’s still to come.
Hey, thanks for reading! If you come here often, and you like what I do, would you please consider contributing something (sorry, that page is a work in progress) to keeping this place running and me out of debtor’s prison? Thank you!
4 thoughts on “Islamic History, Part 6: The problem with early sources”
Can I ask – how did Muhammad take the flight to a mosque in Jerusalem before the religion was formed?
I’m not even sure how to begin answering that. If you actually believe that a man flew from Mecca to Jerusalem on a horse and then visisted Heaven in the year 621, you’re probably not all that interested in nuanced theological arguments about the development of Islam. Even if you are, I’m not sure what one has to do with the other. I don’t know of anybody who argues that al-Aqsa Mosque (I assume by “mosque” you mean the physical, built structure) was already there before Muhammad supposedly made his journey.
I realise it was a rather dumb question and do appreciate your response. I really enjoy your blog and am interested in the nuances of the developments of the Abrahamic religions. I just had to get that one off my chest and probably did so in a bungled way. Fascinated by the historicity of myths, legends and the foundation of religions. Thank you.
Some myths are harder than others to reconcile with history. This seems to me to be one of them.