Mecca before Islam: 6) Ananias of Shirak

The sixth in a series on pre-Islamic sources adduced for Mecca.
Previous: 5) The Samaritan Asatir


ew z-Fa[ṙ]anitē yorum Faṙan awan, zor karcem ’i Tačkac‘ Mak‘a koč‘el
“…and Pharanitis, where the town of Pharan [is located], which I think the Arabs call Mecca”
—the ‘long recension’, tr. Robert H. Hewsen

ew z-Faṙnitis ur z-tun Abrahamu baǰałen
“…and Pharanitis, which is foolishly called the home of Abraham”
—the ‘short recension’, tr. Robert H. Hewsen


Though I have not seen it adduced in the scholarly literature, I have been asked about the references to Mecca in an Armenian Geography attributed to Ananias of Shirak. The translator, Robert H. Hewsen, endorses Ananias as the author; but dating the work as a whole is tricky, and dating specific information within it can be even harder.

Ananias was heavily influenced by earlier geographies in Greek and Armenian, especially those by Ptolemy and Pappus; it is effectively “a mélange of details drawn from sources ranging from the second through the seventh centuries with interpolations made as late as the eighth” (p. 34). According to Hewsen, the information about the author’s native Armenia most accurately reflects the political situation in the last decades of Sasanian rule, starting with the treaty of 591, which settled the borders between Iran and Byzantium, and ending in 636 with the onset of the Muslim conquests. On that basis, Ananias’ firsthand copy of the Geography was probably composed during the lifetime of Muhammad. However, that original manuscript has not survived; instead we have a long and a short recension that were developed over the following decades. The long recension is fairly conservative, with a smattering of later additions, while the short recension has been thoroughly condensed and reworked. Since both of these recensions were developed under Muslim rule, we cannot rule out the possibility that they were ‘updated’ in places to reflect the new political reality.

In each recension the reference to Mecca falls under the section about ‘Rocky Arabia’, a geographical unit recorded by Ptolemy, which spans the northwest of the Arabian Peninsula plus the Sinai Peninsula. Several placenames in this section are ultimately from Ptolemy, including the region of Pharanitis and its eponymous town, Pharan. Interestingly, though, the long and short recensions of Ananias – in rather different ways – identify Pharan and Pharanitis with Mecca. The long recension names “Pharanitis, where the town of Pharan [is located], which I think the Arabs call Mecca.” On the other hand the short recension does not mention Pharan, but refers to “Pharanitis, which is foolishly called the home of Abraham”.

Pharan is, of course, nowhere near Mecca. Their identification is a matter of biblical exegesis. According to Genesis 21, Hagar and her son Ishmael went to live in the desert of Paran. In Late Antiquity it was widely held that the peoples of Arabia were descended from Ishmael; whatever they called themselves, Arabians were often called Ishmaelites or Hagarenes by outsiders. And of course in the Qur’an it is Ishmael, with his father Abraham, who built the “first temple”, which is universally identified in Muslim scholarship with the Ka‘bah at Mecca. Within the logic of exegesis it made sense to locate Mecca within the Desert of Paran. So when Armenian writers encountered ‘Pharan’ in Ptolemy’s ‘Rocky Arabia’, they naturally assumed that this was biblical Paran, which the Muslims had identified with Mecca. If they had known Arabian geography as well as the Muslims did, they might have noticed that Pharan was a very long way from Mecca; but they did the best they could with the resources they had.

These recensions are probably the earliest known writings that associate Mecca with Pharan. And if it was Ananias himself who made this connection, writing before 636, that would constitute an extraordinary piece of evidence: it would mean that Armenians had heard of Mecca and knew about its mythological importance before the Muslim conquests, and perhaps even before Muhammad’s career took off. Are we justified in attributing this small but potent detail to Ananias? Or is there reason to suspect that this detail was added to the Geography as it underwent subsequent revisions, nextdoor to the burgeoning Muslim empire?

The influence of Muslim rule is generally subtle: for instance, the word for parasang, a measure of distance, is given as p‘arsax (p. 43), which echoes the Arabic farsakh more than Greek or Iranian forms of the same word. Thankfully, though, there is hard evidence that towns were added to the Geography after the Muslim conquests. In its description of Iraq, the long recension names “Akałałi, site of the encampment of [the army of] the Arabs”. Clearly this is the town known in Aramaic as ‘Aqula, where Muslim colonists founded the garrison town of Kufa. This detail must have been added to the text after the early stage of the Muslim conquests. The short recension also puts Kufa in Iraq, which is said to have “four districts: Akoła, Basra, Babylon and Ctesiphon.” The latter two were great imperial towns of antiquity, while Basra and Kufa both came to prominence under Muslim rule, serving as administrative capitals for the conquered territories in Iraq and Iran; Ananias would have had no reason to mention Basra and Kufa in his original work.

It is possible that this information in the (later) short recension was expanded from the information in the (earlier) long recension; but the two quite different spellings of ‘Aqula and the utterly different phrasing of the two passages may suggest that these are separate, independent interpolations. In this case, two editors working on different versions of the Geography each noticed that the section on Iraq was a little outdated, and each decided to bring the text up to date by adding the name of a newly prominent Muslim town or two.

Under the circumstances, we really ought to suspect that the references to Mecca were late additions to Ananias’ text, just like Basra and Kufa. In each recension the comment about Mecca is parenthetical, glossing Pharan and Pharanitis without changing substantially the flow of the text. Moreover, the phrasing is so distinct between the two recensions that we may wonder if these, too, are independent additions. One explains straightforwardly that Pharan is called Mecca by the Arabs; the other expresses the same notion in euphemistic terms, that Pharanitis is “foolishly called the home of Abraham”, without deigning to tell us who believes this. But even if we take for granted that these two statements have a common origin, there is no compelling reason to think this was part of Ananias’ original work. There is, after all, no supporting evidence that Mecca was known to Armenian writers or their neighbours before Islam; and it would make perfect sense for editors to insert a reference to Mecca, along with Kufa and Basra, under the influence of Muslim hegemony.

Ananias’ Geography would not be alone in this. A Middle Persian geography called The Provincial Capitals of Iran seems to have taken shape under Sasanian rule, but was further developed under Muslim rule. Among the cities of the Near East it lists Mecca as well as Kufa, Mosul, Medina, and Baghdad, all of which were probably added some time after the Muslim conquests. As we have seen, Kufa came to prominence as a Muslim regional capital; so did Mosul. The text itself acknowledges that Baghdad was founded by the caliph al-Mansur. The oasis of Yathrib only became known as Medina – ‘the city’ – after Muhammad established his polity there. Likewise, the text uses Arabic geographical terms for Syria, the Yemen, North Africa and upper Mesopotamia. Such is the extent of this tampering, we are forced to assume that Mecca too was a later addition. The Provincial Capitals may in other respects be a useful source for Sasanian geography, but it can hardly serve as evidence for Mecca’s importance before Islam. The same, unfortunately, must be said for Ananias.



[the long recension] Arsène Soukry (ed. and tr.), Géographie de Moïse de Corène (Venice: Imprimerie arménienne, 1881), Mecca on 45 (French) = *37 (Armenian).

[the short recension] Jean St-Martin (ed. and tr.), Mémoires historiques et géographiques sur l’Arménie, vol. 2 (Paris: Imprimerie royale, 1819), Mecca on 368–369 (Armenian and French).

Robert H. Hewsen (tr.), The Geography of Ananias of Širak (Ašxarhac‘oyc‘): The Long and Short Recensions (Wiesbaden: Ludwig Reichert, 1992), Mecca on 70–71, 70A–71A [link].

Tim Greenwood, “Ananias of Shirak”, in the Encyclopædia Iranica [link].

Touraj Daryaee (ed. and tr.), Šahrestānīhā-ī Ērānšahr: A Middle Persian Text on Late Antique Geography, Epic and History (Costa Mesa, CA: Mazda, 2002), Mecca on 15 = 19 (§33), and commentary on 46–7.

Touraj Daryaee, “Šahrestānīhā ī Ērānšahr”, in the Encyclopædia Iranica [link].

Ian D. Morris

Ian D. Morris

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

Leave a Reply

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