For well over a millennium Muslims have revered Mecca as the site of their holiest shrine, the Kaaba. And until recently Western scholarship always accepted the traditional Muslim origins narrative, which says that was where Muhammad began, in Arabia. But in the late 1970s, John Wansbrough and two of his students, Patricia Crone and Michael Cook, published books arguing for a radically different approach to Islam’s origins.[i]

Among other things, these revisionists said that Mecca was not Islam’s birthplace, which they located somewhere in the Fertile Crescent. Though Crone and Cook later repudiated the theory their book advanced, Crone at least held fast to the idea that Islam originated in the Fertile Crescent, possibly in Nabatea.

While the trend among Western scholars of Islam is away from such radical doubt, four decades later some scholars still promote the idea that the Kaaba was not originally in Mecca.

Some of these revisionists say it was in or near Petra, while others refuse to speculate on the exact location. And this notion has begun to trickle down to others in the West, through the work of popular historian and documentary filmmaker Tom Holland, for example.[ii] Since Muslims pray facing Mecca’s Kaaba multiple times a day, one thing this view would mean is that Muslims everywhere naively face the wrong direction in their most frequent act of worship.

Revisionists variously claim the following evidence supports their theory:

• The qur’anic data
• Other early written evidence
• The hadith’s data on Mecca
• The archaeological record
• Al-Tabari’s historical record
• Mecca’s geographic conditions

It’s important to recognize that ancient history rarely offers the absolute proof of an airtight case. Instead, we must base our conclusion on the preponderance of the evidence.

Besides these six lines of evidence, we must also consider how plausible it is that the early Muslim community reassigned its origins to a different city than that of its actual birthplace. Most revisionists allege that Muslims made this change during Islam’s classical period for political reasons. Yet how believable is it that Islam originated not in Mecca, but in Petra or somewhere else in Nabatea?

1. The qur’anic data

Regarding the qur’anic evidence, the fact that the Qur’an names Mecca just once[iii] may look suspicious when compared with the New Testament’s naming Jerusalem, for example, almost 180 times. However, the two scriptures are radically different books. By way of comparison, the Qur’an names only a few contemporary geographic locations and none more than a couple of times, while the New Testament names many towns and other geographic features numerous times. Likewise, the Qur’an names Muhammad just four times, while the New Testament uses the name Jesus well over a thousand times. Thus, the important point with reference to the Qur’an’s naming of Mecca is that it names it in relation to its own story,[iv] whereas Petra (al-Raqim) is mentioned only in relation to a historical event.[v] And there’s nothing about that mention to suggest that al-Raqim is in the vicinity of Muhammad’s hearers.

Some scholars think other qur’anic data point to Islam’s having originated in Nabatea. They cite

• Sodom’s location in relation to Muhammad’s hearers
• Animals mentioned in the Qur’an
• Fruits mentioned in the Qur’an

But much here depends on how we interpret the text. A literal reading of Q 37:137 locates Muhammad’s hearers close enough to Sodom’s Nabatean ruins that they passed by them twice daily. (But even Petra is ruled out by a literal reading since Petra is some 80 kilometers or 50 miles distant from the south end of the Dead Sea, where Sodom’s ruins are supposed to lie. And a return journey of that distance wasn’t been possible back then.) A freer reading allows for the traditional interpretation. It puts those ruins in the vicinity of a caravan route to Syria many of Muhammad’s hearers would have been familiar with.

Q 80:24-32 and other Meccan passages speak of God’s provision of fruit and anʿām—sometimes translated “cattle”—which Mecca’s climate would not have allowed. But anʿām can also be translated “beasts,” which could mean camels. This would make the verse’s provisions “for you and your beasts” relevant to traders and camel herders alike. Q 6:136-139 implies that Muhammad’s opponents were themselves farmers. But such passages may well detail practices of residents of pagan Ta’if, just 87 kilometers (54 miles) from Mecca. In fact, most of the Qur’an’s agricultural references require a locale no further afield than Ta’if, famous for its grapes, pomegranates, figs, etc. Only olives would require the growing conditions found in Yemen and Nabatea-Palestine, at either end of western Arabia’s overland trade route. Thus, almost everything in the Qur’an’s early suras is compatible with its Hijazi origins. There is also no reason to reject the traditional Muslim view that these texts speak universally (possibly after the pattern of the biblical psalms[vi]).

Revisionists must reckon with two other facts which argue against Islam’s Nabatean origins:

• The presence of 200 Amharic and Ethiopic loanwords in the Qur’an[vii]
• Qur’anic references to pagans’ idol worship and animal sacrifice

Significant linguistic borrowing suggests extensive cross-cultural interaction. When goods and ideas are exchanged, words often are as well. Cultural dominance may play into linguistic borrowing also, and Ethiopia ruled the Hijaz for a time during the 6th century. If the Qur’an’s early suras were addressed to the inhabitants of Petra, one might expect more Coptic than Amharic and Ethiopic loanwords since Nabatea was much closer culturally to the Graeco-Roman world, Egypt included, than to Ethiopia. Yet Amharic and Ethiopic words in the Qur’an stand in a 20:1 ratio to Coptic words.[viii] While this simple numerical comparison is not conclusive, it certainly raises questions.

Regarding pagan practices, the Byzantines had forbidden both idol worship and animal sacrifice long before Muhammad’s time—including in their province of Arabia Petraea.[ix] Yet the Qur’an repeatedly refers to idolatry as a contemporary practice, calling the unbelievers to forsake their idols, which they look to for protection (e.g., Q 2:256-57, 16:36). G.R. Hawting has argued that Muhammad challenged only the “spiritual idolatry” of retrograde monotheists.[x] But in its listing of proscribed foods, Q 5:3 says, “Forbidden to you are carrion, blood, pork…  whatever has been sacrificed to idols.” This was clearly pagan idolatry, which points to a region like Arabia’s Hijaz, beyond the bounds of the Byzantine Empire. Q 22:30 also warns against contamination by “the filth of idols” (awathan). The Qur’an also condemns “sacrificial stones” and the meat sacrificed on them (Q 5:3, 5:90). But since sacrifice had long ceased to be part of Jewish practice and was never practiced by Christians, this can only relate to idolatrous practice. The Qur’an repeatedly bans “that on which any name other than God has been invoked” (e.g., Q 2:173, 5:3). Likewise, Abraham is repeatedly presented as the prophetic hero who challenged his people’s idolatry (e.g., Q 26:69-102) as Muhammad is now doing.

How does such paganism square with the Qur’an’s mention of the idolaters’ belief that God (Allah) created the world? In fact, it’s consonant with what we know of widespread polytheistic belief in a High God.[xi] Q 6:136 even describes the pagans as offering some portion of their produce to God, while other passages present them as ascribing offspring to God, swearing by God and even praying to God when in distress (e.g., 6:63-64, 6:109, 16:57). Not only is this qur’anic picture of pre-Islamic paganism consonant with polytheistic belief in a High God. It’s also generally consistent with both pre-Islamic poetry and the Book of Idols. And epigraphic evidence from both Palmyra and South Arabia attests to the pre-Islamic Arabs’ ascription of daughters to God. To sum up, the Qur’an condemns not adulterated monotheism, but rather literal idolatrous polytheism, something that had by the early seventh century been long forbidden in Byzantine Nabatea.[xii]

 Thus, the Qur’an’s mention of Mecca, its Ethiopic loanwords, and its references to idolatrous practice all make the Hijaz a location more likely for Islam’s emergence than Nabatea.

2. Other early written evidence

Some scholars believe the lateness of the earliest written extra-qur’anic evidence for Mecca renders it unreliable. Revisionists clearly reject the idea that Ptolemy included Mecca, as “Makoraba,” on his second century CE map of Arabia. That question is of relatively minor import. But regardless, they thus conclude that the first map documenting Mecca’s existence is late.

Regarding textual evidence, only a small percentage survives from any ancient culture. And unlike the Mediterranean world at the time of Christ, Arab culture was oral during Islam’s first two centuries, producing little written Arabic before the ninth century, beyond inscriptions, graffiti, business and administrative records, and the Qur’an. Hence, the earliest extra-qur’anic Arabic mention of Mecca comes from the late seventh century, with most of the early mentions being from the late eighth century. However, wishing we had earlier evidence doesn’t license us to discount the early Arabic evidence we do have. And all the early written accounts of Islam’s origins point to Mecca, none to Petra or Nabatea.[xiii]

3. The hadith’s data on Mecca

We should take hadith descriptions of Mecca’s grandeur and lush vegetation as hyperbole designed to glorify Mecca, like the Qur’an’s designation of its home base as the “Mother of Settlements” (Q 6:92, 42:5). Crone is doubtless right to argue that western Arabia’s economy was unable to support the populations mentioned in the hadith.[xiv] Neither was Mecca on a major trade route. But again, we should not allow the Qur’an’s hyperbolic designation or the hadith’s hyperbolic elaborations to mislead us into looking for a large city at the nexus of a trade empire.

Hadith sources consistently disagree when hyperbolizing. It is where they consistently agree that we should pay attention. And they invariably make Mecca Islam’s birthplace.

4. The archaeological record

Revisionists make two claims about the archaeological record. They claim that

• There’s no archaeological evidence that Mecca was inhabited in the seventh century
• The qibla, or prayer direction, of most of Islam’s early mosques points to Petra, not Mecca

Due to the Saudi Arabian government’s mortal dread of shirk, or polytheism, it has strictly forbidden all archaeological study of Mecca’s historic sites, lest it lead to relic worship. Indeed, the Saudis seem determined virtually to obliterate the city’s archaeological record in their rush to ring the Kaaba with skyscrapers. An estimated 95% of Mecca’s historic buildings have been demolished to allow for this building spree. Any remaining historic sites are treated with a combination of fear, contempt and scholarly avoidance, lest they be idolized.[xv]

Unfortunately, this leaves us with no archaeological evidence either for or against Mecca’s being Islam’s birthplace.

 The other archaeological claim some revisionists make relates to early mosque orientation. Following Crone, writer and documentary filmmaker Dan Gibson claims close agreement in the qibla, or prayer direction, of most of Islam’s early mosques—but to Petra, not Mecca.[xvi] Some enthusiasts of Gibson’s view claim that the reliability of his assertions is easily demonstrated simply by using the satellite imagery of early mosques accessible through Google Earth. But any confirmation this tool yields is unreliable due to the fact that many early mosques have multiple foundations, representing various qibla orientations. In other words, we can only prove an ancient mosque’s original orientation by conducting a thorough archeological study of these mosques, digging down to their first foundations. While Crone quickly accepted this correction, Gibson has not yet done so.

The level of qibla agreement Gibson claims to have found is also impossible for a number of reasons. To begin, their builders used conflicting methods to determine the qibla, just as American Muslims do today.[xvii] Reflecting these conflicting methods, the mosques they built do not agree, though all their builders did their best to orient them to Mecca. In addition, the early Muslims had a very basic understanding of geography. As Islamic science historian David A. King explains, “The first generations of Muslims had no means whatsoever for finding the direction of Petra [or Mecca either] accurately to within a degree or two, not least because they had no access to any geographical coordinates, let alone modern ones, and no mathematics whatsoever.” The early Muslims determined the qibla accurately by the standards of the day, using the best folklore-based methods at their disposal.[xviii]

But with only primitive astronomy and maps and no mathematics, they were unable to achieve anything close to modern-day accuracy.[xix] Thus, while the early Muslims oriented their mosques toward Mecca, their limited abilities and conflicting methodologies makes this far from self-evident.

5. Al-Tabari’s historical record

Reading between the lines, Gibson suggests that al-Tabari’s account of Ibn al-Zubayr’s trip to Mecca in 70 AH (689-90 CE) may point to the Muslim community’s relocation from Petra to Mecca.[xx] Tabari says Ibn al-Zubayr took “many horses and camels and much baggage” with him to Mecca. But had the rebel Ibn al-Zubayr’s trip represented a communal move and a relocation of the Black Stone to Mecca, why would his enemies not have reversed it upon his defeat? As for the horses mentioned, he would have needed them to mount the defense of his chosen refuge. Ibn al-Zubayr needed the money since transferring power from Damascus involved outfitting and rewarding his supporters, and money (all coins) was heavy in those days. Tabari also says that many camels were slaughtered on his arrival in Mecca—doubtless to celebrate his victory, fleeting though it was. There’s nothing to suggest that this points to the Muslim community’s relocation of Islam’s holiest shrine.[xxi] And while Tabari never once mentions Petra, he elsewhere repeatedly names Mecca as home of the Kaaba.

6. Mecca’s geographic conditions

Some argue that Mecca’s harsh conditions and geographic isolation make it a wretched choice for the spiritual center of the world.[xxii] But Muhammad never claimed to choose Mecca. Rather, Mecca chose him. And however ambitious he was, it seems likely that he initially hoped to make his hometown simply the center of his Arabian theocracy. When he first began, he could not have known how much of the globe his armies would subdue.

7. The question of plausibility

The last issue for us to consider in assessing the theory that Petra or some other city in the Fertile Crescent is Islam’s real birthplace is that of the plausibility. That is, how plausible is the move the theory necessitates from Islam’s alleged birthplace to Mecca?

Specifically, how could the Muslim community have seamlessly made and accepted the move from Islam’s “real birthplace” to Mecca, its “pseudo-birthplace,” without leaving any trace of that move in the written record?

Of the proposed answers to this question, two call for our consideration. They are that

• Muslims called Petra “Mecca” prior to the alleged move
• The early Muslims determined to cover their tracks

Gibson puts forward the hypothesis that Muslims formerly called Petra “Mecca,” a hypothesis endorsed by Christian apologist Jay Smith. This would mean there were two Meccas, the first being Petra, the second being the Kaaba’s current home in Saudi Arabia. Thus, when the Qur’an names Mecca, it’s referring to the original Mecca—that is, Petra. Likewise, the hadith accurately describe the Nabatean “Mecca” before the Muslim community gave the Kaaba’s new Hijazi home the same name. What evidence does Gibson have to support his claim that “Mecca” originally referred to Petra? He bases it on the testimony of a 9th-10th century Christian historian named Thomas Artsruni. He wrote that Muhammad had preached in Mecca, located in “Arabia Petraea Paran.” According to Gibson, Thomas locates Mecca in Petra, “in southern Jordan.”[xxiii]

However, Thomas locates Mecca, not in the city of Petra at all, but only in the Byzantine province of Arabia Petraea, specifically in its Paran, or Sinai, region. Two things explain Thomas’s mistake. First, he wrote in his native Armenia and wrote of distant places he’d never seen. Second, he undoubtedly placed Mecca in Paran because Muslims claim that Mecca was the site of Hagar and Ishmael’s exile, an event Genesis 21:22 locates in Paran. In other words, Thomas mistakenly assumed that Mecca must be in Paran since the author of Genesis set Hagar and Ishmael’s story there.[xxiv] Thus, we should overlook Thomas’s error, not build on it.

The other explanation for the hadith’s total silence on the topic of the Muslim community’s alleged move of its central shrine is that the early Muslims determined to cover their tracks. They wanted their move of the qibla quickly forgotten, lest it diminish Mecca’s sanctity and legitimacy. But the realities of the period in which the alleged move occurred argue strongly against the hadith’s complete failure to mention it:

• The Muslim community was widely distributed within a few years of Muhammad’s death in 632 CE.
• The community was deeply divided from that point on, with Sunni, Shia, Khariji and other Muslim groups fighting to gain or hold onto power.
• The alleged change of qibla would have been a matter of paramount importance since Muslims have always worshipped in a single prescribed direction.
• The hadith reflect the early Muslim community’s regional, sectarian and other divisions and disagreements in numerous respects.[xxv]

Yet, despite the community’s broad geographic distribution and deep division and the central importance of the qibla, the hadith register no disagreement whatsoever on this point. To think that so sprawling and so unruly a community could either unanimously agree to relocate its sacred center or somehow manage to do so without leaving a single trace of the move in the hadith record is highly implausible to say the least.

To sum up, ancient history seldom offers the sort of proof that enables us to build an airtight case. Thus, we can only carefully weigh all the evidence and decide which conclusion most of the best evidence supports. As I have shown here, none of the six lines of evidence revisionists use to support their view is compelling. Instead, the preponderance of the evidence points to Mecca as Islam’s true birthplace. That is especially true when we allow that oral history—like that later recorded in the hadith—is not inherently false and consider the implausibility of the alleged relocation of the Kaaba.

Hence, contrary to what most revisionists claim, we can affirm with reasonable confidence that Muslims do not mistakenly face the wrong direction when they pray.