By Kyle Orton (@KyleWOrton) on 9 June 2025

The “Constitution of Medina” or “Medina Document” (Sahifat al-Medina) is recorded in the Islamic Tradition literature, which post-dates the ninth century, i.e., two-hundred years after the Prophet Muhammad died, but most modern historians accept the Tradition’s claim that the Sahifa is the authentic record of a covenant between the early Muhammadan community and the Jews of Yathrib to establish governance in the city, dating from around 622 AD. This view is taken because the Sahifa is linguistically distinct from the surrounding material and, crucially, it contains elements that are theologically awkward from the perspective of “final form” Islam that stabilises after 800 AD.

Perhaps the most awkward aspect of the Sahifa is that it includes some of the Jews within the umma (community) of believers. As problematic, the Sahifa enumerates six Jewish clans involved in the treaty with Muhammad (who is named), but the Banu Qaynuqa, Banu Nadir, and Banu Qurayza are nowhere to be found. In the Tradition, Muhammad aligns with these three Jewish tribes at Medina and they break the pacts, leading to the expulsion of the first two and the harrowing of the last, with its men slaughtered, and women and children sold into slavery. The terms of the polity described are immensely tribal, including divisions of responsibility and commitments to pay in “matters of ransom or blood-money, as is customary”. In other words, it does not describe an Islamic State: Muhammad seems more like the manager of a tribal confederation than a caliphal leader; there is no mention of the shari’a; the Jews are not treated as dhimmis; and the word “Muslims” appears only three times, each of them somewhat strange from the later vantage point. In one case, “Muslims” are even separated from “Believers”.[1]

As strange is what is not in the Sahifa: there is no narration of the hijra or exodus that Tradition says brought Muhammad’s community to Yathrib, and the city is named throughout as Yathrib, not Medina, which is favoured in Islamic orthodoxy. (The word “medina” does technically appear in the Sahifa, but it is in passing when referring to security arrangements, not as a name: “… whoever remains in place is safe in the city (al-medina) …”.) It leaves a question about the identification of Yathrib and Medina.

The Tradition has it as an assumption that Yathrib and Medina are one and the same, and locates the city where we know it today in Saudi Arabia. But the Qur’an, which unlike the Tradition seems to be genuinely Late Antique, does not say this. Yathrib is mentioned once in the Qur’an (33:13) and it has no geographical markers attached to it; the mentions of “al-medina” are in the format indicated above, to mean “the city”. Still, even most Revisionist historians in Islamic Studies accept the basic premise that the Arab conquests, starting with the Syria-Palaestina zone, were launched from the Hijazi Medina, about 550 miles away.

Patricia Crone and Michael Cook notoriously repudiated the “central thesis” of Hagarism: The Making of the Islamic World (1977), yet the book was not without its insights, one of which was that while Muhammad’s “base [for the invasions] could conceivably have been Yathrib [in the Hijaz], … general geographical plausibility might suggest a location farther north” (p. 24). The broader surrounding evidence that Muhammad’s community originated in the “sectarian milieu” of the Roman Levant, and the possibility that the collapse of Roman rule resulted more from internal implosion after the long war with Persia, buttress the suggestion.

It is, therefore, notable that the first non-Tradition source to securely identify Yathrib and Medina as synonymous, the Syriac Khuzestan Chronicle in the 650s, does so in this way (p. 31): “The city of Hazor, which Holy Scripture [Joshua 11:10] calls the head of kingdoms, also belongs to the Arabs, and is called Medina [by them], from the name of Midian, the fourth son of Abraham, whom he had begotten from Keturah. The city is also called Yathrib.” Hazor is in the Galilee, near what is now the Golan Heights in northern Israel. Notable, too, the people of Midian come to grief in the Qur’an (7:85-93) as one of the nearby peoples who reject Muhammad, and Midian was known for centuries before the advent of Islam to be located around the Gulf of Aqaba, opposite the Sinai, in what is now southern Jordan.[2]

The chronicle may be in error, but it is near-contemporary and it points to the same setting in and around the Biblical Holy Land for Muhammad and his followers that so much of the other evidence in the period indicates.

There are two other interesting sources on this, though they are less important because they are later.

Thomas Artsruni, an Armenian writing in the late ninth century, explicitly identified Mecca in his History of the House of Artsrunik with Pharan (or Paran), a town in what had been Arabia Petraea, the Ghassanid territory within the Empire’s borders in the Roman Levant that had Petra and then Bosra (now southern Syria) as its capitals.[3]

The Chronicle of Seert, compiled in Iraq in the tenth century, though including a lot of material back to the seventh century if not earlier, by a Christian from the Church of the East (hence sometimes called the Histoire Nestorienne), reports: “[Muhammad] made his abode in Yathrib, which is the city of Keturah, the concubine of Abraham, and he called it Medina.” It stands out because the Seert/Si’ird Chronicle author seems otherwise to have only information from the by-then-stabilised Islamic Tradition.[4]

---

[1] Fred Donner (2010), Muhammad and the Believers: At the Origins of Islam, pp. 227-32.

[2] James K. Hoffmeier (2005), Ancient Israel in Sinai: The Evidence for the Authenticity of the Wilderness Tradition, p. 121.

[3] Ian D. Morris (2018), ‘Mecca and Macoraba’, Al-Usur al-Wusta. Available here.

[4] Philip Wood (2013), The Chronicle of Seert: Christian Historical Imagination in Late Antique Iraq, pp. 244-245.