Mecca before Islam: 3) Dabanegoris regio

The third in a series on pre-Islamic sources adduced for Mecca.
Previous: 2) Macoraba.
: 4) Narsai’s homily.

a flumine Canis, ut Iuba, …gens Taludaei, Dabanegoris regio, mons Orsa cum portu…
“After the Dog’s River, according to Juba, there is… the Taludaei
tribe, the Dabanegoris district, Mount Orsa with its harbour…”

—Pliny the Elder, tr.  Harris Rackham


For students of ancient Arabia, Pliny the Elder (d. 79) is indispensable. A military officer and administrator in the early Roman Empire, he was also a scholar of varied interests. His works on Roman history, military tactics and political rhetoric are lost, but his one surviving work is also the longest of all surviving texts in Classical Latin: a Natural History that meant to describe the whole of nature and the many ways we encounter it. This encyclopaedic ambition led him to write long sections on geography,  which have been foundational sources for historical geographers since the Middle Ages. Pliny’s description of the Arabian Peninsula is cogent and sometimes corroborated by other ancient writers. There is no reason to think that Pliny himself ever did fieldwork in Arabia; he relied on earlier studies, often citing the popular works of Juba II.

Before ascending the throne of Mauretania in the Maghreb, Juba (d. 23) was a bookish young exile in Rome who wrote on topics as diverse as archaeology, linguistics, painting and music. He was not himself a Mauretanian, but he was descended from a royal line in the neighbouring kingdom of Numidia, which was a qualification of sorts; and to seal the deal he was married to an Egyptian princess, the daughter of Cleopatra and Mark Antony. Those African credentials were enough for the Roman Empire to install him as a friendly monarch. Juba then applied his scholarly powers to mapping out his new realm, from the Atlantic to the Atlas, and its easterly neighbours as far as Egypt.

Juba’s studies on North Africa established him in the Roman intellectual sphere as a leading authority on the southern half of the known world. Augustus therefore recruited his expertise to an expedition to the eastern frontier of the Empire. Consulting local merchants and intellects, as well as the libraries of Alexandria and Kappadokia, Juba wrote his next major work, a study of the trade routes passing through Arabia to India. Juba did not himself travel through Arabia, but his description of the Peninsula became one of the major sources for other classical writers, especially Pliny, who was born around the year Juba died.


Among the places in Juba’s description of Arabia is Dabanegoris regio, ‘the region of Dabanegoris’. In 1970 the geographer Hermann von Wissmann proposed that this was the territory belonging to the Quraysh: Mecca and its environs. Dabanegoris would then mean literally ‘that which belongs to the Sons of Quraysh’.

This derivation could not be Arabic; in fact dhū banī Quraysh would mean entirely the opposite, ‘that which owns the Sons of Quraysh’. Rather, Wissmann had derived it from Ancient South Arabian, a family of languages that dominated the Yemen in the centuries before Islam. In that case, Dabanegoris would be something like DBN QR(Y)Š. Such a construct would be grammatically correct, but Wissmann did not explain why a region in western Arabia should be known by a South Arabian name. And that is the least of its problems.

First of all, Dabanegoris will have been shifted by the rules of Latin grammar. Latin nouns carry grammatical information by changing how they end, as many schoolchildren know. In the term Dabanegoris regio, the word regio ‘belongs to’ the word Dabanegoris, so we should expect to see the latter in the genitive case; and happily, the ending –oris would suggest that this is a singular genitive noun in the so-called third declension. If we reverse-engineer Dabanegoris into the nominative case, so the word can stand alone, we should expect to find a rather different ending: perhaps *Dabanegos. You can see the problem: Wissmann read Quraysh into Dabanegoris, but that –oris is a declension, not an integral part of the root word. If Dabanegoris regio is ‘the region of Dabanegos’, there is no room left for the Quraysh.

Moreover, as Patricia Crone has observed, the term ‘Sons of Quraysh’ (Banū Quraysh) is a bit faulty. ‘Quraysh’ is the name of the tribe, not a patriarch within the tribe, so the patronymic ‘Sons of Quraysh’ would have made little sense to anyone. In practice, if the Quraysh wanted a patronymic, they would call themselves the ‘Sons of Fihr’, because Fihr was a patriarch. Wissmann needed the element banū to make up Dabanegoris, but there is no good reason for it to be there.

Even if we did approve Wissmann’s reading of the name, the details of geography would still be against it. Pliny, citing Juba, follows the Arabian coast from Charax down the Persian Gulf, around Oman, dropping the name Dabanegoris regio along the way. Juba does admit that his sources have not fully navigated the coastline, which is too rocky, but enough details are corroborated by other ancient sources that we can be reasonably confident in his accuracy. It is clear that Dabanegoris regio belongs in the southeast of Arabia, not the west. Wissmann was not the first to misread this passage, for whatever reason; but we should not repeat the mistake. It is evident that Dabanegoris regio has nothing to do with the Quraysh, and even if it did, the region could not possibly include Mecca.


We might also question whether a people by the name of Quraysh could have existed by Pliny’s time, let alone Juba’s. The tribe’s early history is admittedly sparse, drawn as it is from the oral traditions of early Muslim society. From these we might conclude that the tribe first took the name Quraysh after conquering Mecca in the fifth century, give or take; far too late for Dabanegoris regio. Then again, origin stories are famously unreliable: it is possible that the name Quraysh had deeper roots than the semi-legendary material that medieval scholars were writing and codifying by the ninth century.

Wissmann believed he could find the Quraysh in the historical record a couple of centuries earlier than the oral tradition would allow. The kings of Hadramawt in the Yemen used to hold ceremonies at al-‘Uqlah, where they would have commemorative notes inscribed on the surrounding rocks. One inscription (Ja 919 / RES 4862) records that Il‘azz Yaluṭ, son of ‘Ammzakhar, was visited by a group of thirteen women with Arabian names, affiliated with a people or place written as QRŠ. Albert Jamme, who published this inscription in 1963, posited that these women were from the Quraysh; Wissmann agreed.

It is at least plausible that Yemeni kings should have been visited by foreigners with a stake in the incense route, which ran from India, via the Yemen, into the empires of the Near East. Another inscription records that the same Il‘azz Yaluṭ was visited by persons from India as well as Palmyra (Syria) and Chaldea (Iraq), all of whom were presumably connected by this route. If Palmyrenes and Chaldeans had a presence at the Yemeni courts, there could easily have been delegates from elsewhere in Arabia.

Il‘azz Yaluṭ reigned, it seems, in the early third century. This is not early enough to corroborate Wissmann’s reading of Dabanegoris regio, which now seems to have failed on all fronts: even with his Ancient South Arabian evidence, the Quraysh would only be attested 150 years after Pliny and 200 years after Juba. And if we do allow the possibility that the QRŠ are Quraysh, we are faced with a new problem: why should the Quraysh pop into the historical record in the early third century and then vanish for centuries to come?

Thanks to the oral tradition in Arabic, we can be reasonably confident that the Quraysh were extant and settled in Mecca by the sixth century; but no contemporary source mentions them. Unlike the kings of Yemen and Palmyra, the Quraysh left no trace in their neighbours’ writings; their earliest guaranteed entry in the historical record comes after the Muslim conquests, when this relatively obscure tribe was propelled to imperial glory. In the 400 years between the reign of Il‘azz Yaluṭ and the prophetic career of Muḥammad, the Quraysh are nowhere to be seen.

This is a problem. If we had an inscription like the one at al-‘Uqlah, but with a later date – say, around 550 – we could more easily believe that the QRŠ were the Quraysh that we know from a slightly later time. If we then found another dating to 500, and another to 400, both mentioning a people called the QRŠ, we could then proceed on the assumption that these are a continuum, paving a road for the Quraysh going back through the centuries. But all we have is the inscription of Il‘azz Yaluṭ – a farflung outlier – and then silence through Late Antiquity. It may be easier to conclude that the QRŠ are not Quraysh, or at least, that such an identification would be highly speculative and reckless to build on.

The Quraysh were minor players on the world stage until the lifetime of Muḥammad, and Mecca was a minor settlement; it would be quite a surprise to find them in Pliny, half a millennium before Islam.


Further reading

Hermann von Wissmann, “Makoraba”, Supplement to Paulys Realencyclopädie, vol. 12 (1970), 792.
Patricia Crone, Meccan Trade (1987), 134–5, 169.
H. Rackham, Pliny: Natural History, vol. 2 (1961), 448–51 (§6.32).
Duane W. Roller, The World of Juba II and Kleopatra Selene (2003).
Albert W.F. Jamme, The al-‘Uqlah Texts (1963), 37–9.

Ian D. Morris

Ian D. Morris

Historian and occasional human, powered by tea and cat videos.


  1. May be you should look for Quraysh in Petra not Mecca since that’s where they were living back then.

  2. First thanks to do this work Ian.
    Yawn. It seems that “Mecca” is still missing. Nihil nove sub sole..”Mecca” (I mean the Mecca of Ibn Ishaq of course,the only one that matters, capable to raise 50,000 strong and to feed them -with stones maybe…- to do the jihad against the Romans and the Persians…) did not exist before Islam…

  3. The key here is not Dabanegoris but maybe the Dog’s river… Many kuraish tribes are banu kalb (sons of dog??)

    • I don’t see how someone could argue that, but also, I think there’s a broader lesson to be drawn from this blog series. Scholars have tried to project features of Early Islam back into Classical Antiquity. In order to find Mecca and the Quraysh in ancient sources, we have resorted to weak and contradictory arguments. We should ask ourselves why these arguments have predominated; what are our motivations? If we can understand that, maybe then we can resist the temptation to grasp at straws.

Leave a Reply to lolo Cancel reply

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