+++
title = "Petra"
slug = "petra"
description = "Petra (Greek Πέτρα 'rock'; Nabataean Aramaic רקמו Raqmū; Hebrew רֶקֶם Reqem; Arabic ٱلْبَتْرَاء al-Batrāʾ) is the rock-cut capital of the Nabataean kingdom in present-day southern Jordan, and the geographic centre of a revisionist hypothesis — associated principally with the Canadian researcher Dan Gibson — that holds Petra, not the Hijazi city now called Mecca, to have been the original holy city and qibla-focus of early Islam. The mainstream-uncontroversial facts are that Petra was the caravan-trade capital of the Nabataeans from roughly the fourth century BCE, was annexed by Rome in 106 CE as part of Provincia Arabia, retained a mixed pagan-and-Christian population into the Byzantine period, and was abandoned as a major centre after the earthquakes and trade-route shifts of late antiquity. Gibson's contested archaeological argument, developed across Qur'anic Geography (2011) and Let the Stones Speak (2023), reads the orientation (qibla) of the earliest mosques — which on his survey face Petra rather than the Hijaz for roughly the first century of Islamic building — together with the absence of Mecca from Ptolemy's Geography, the irregular four-sided Ka'ba measurements recorded by al-Azraqī, and the Nabataean-Aramaic substrate of Qur'anic Arabic, as converging evidence that the sanctuary, pilgrimage, and Black Stone were relocated south to the Hijaz only in the late seventh and early eighth centuries CE. The hypothesis is rejected by the mainstream historian of Islamic astronomy David A. King and is not part of the Raëlian canon; the Wheel of Heaven framework engages it as a revisionist reading that, if sound, would relocate the geography of the Hanafiyya — the 'religion of Abraham' — to the Nabataean capital, sharpening rather than displacing the corpus reading of early Islam as an Abrahamic restoration movement."
template = "wiki-page.html"
toc = true

[extra]
image = "https://assets.wheelofheaven.world/images/wiki/petra-nabataean-v1.webp"
image_avif = "https://assets.wheelofheaven.world/images/wiki/petra-nabataean-v1.avif"
image_alt = "The rock-cut Nabataean façades of Petra in rose-red sandstone, carved into a canyon cliff under low golden light."
image_caption = "Petra — the rock-cut Nabataean capital at the centre of the qibla-orientation hypothesis."
category = "Places & Locations"
editorial_pass = "2026-05"
entry_type = "place"
claim_type = "speculative"
timeline = ["pisces"]
see_also = [
    { title = "Nabataeans", path = "wiki/nabataeans", description = "The Arab kingdom whose rock-cut capital Petra was." },
    { title = "Hanafiyya", path = "wiki/hanafiyya", description = "The religion of Abraham whose geography the hypothesis relocates to Petra." },
    { title = "Muhammad", path = "wiki/muhammad", description = "The prophet whose earliest community the qibla-orientation argument reorients toward Petra." },
    { title = "Abraham", path = "wiki/abraham", description = "The patriarch whose restored religion the corpus reads early Islam as reviving." },
    { title = "Plurality of Gods", path = "wiki/plurality-of-gods", description = "The Nabataean pagan pantheon Petra hosted before its Abrahamic reading." },
    { title = "Genesis", path = "wiki/genesis", description = "The scriptural Abrahamic backdrop against which the Petra hypothesis is read." },
]
alternative_names = ["Πέτρα (Greek 'rock')", "רקמו (Nabataean Raqmū)", "רֶקֶם (Hebrew Reqem)", "Rekem", "Raqēmō", "ٱلْبَتْרَاء (Arabic al-Batrāʾ)", "Sela (biblical 'rock')", "Mother of the Colonies", "Reqem of the Gorges"]

[extra.infobox]
type = "Rock-cut Nabataean caravan capital; UNESCO World Heritage Site; geographic centre of the Petra-as-original-Mecca hypothesis"
location = "Wadi Musa, Ma'an Governorate, southern Jordan; approximately 1,200 km north of present-day Mecca"
nabataean_name = "רקמו (Raqmū / Reqem), the kingdom's own designation; 'Petra' (πέτρα, 'rock') is the Greek-Roman name"
flourished = "c. 4th century BCE – 2nd century CE as Nabataean capital; annexed by Rome 106 CE (Provincia Arabia); declined after the 363 CE and later earthquakes"
mainstream_status = "Nabataean kingdom capital; major Hellenistic-Roman caravan city; mixed pagan-Christian population into the Byzantine era"
hypothesis = "Dan Gibson's argument (Qur'anic Geography 2011; Let the Stones Speak 2023) that Petra was the original holy city, qibla, and pilgrimage-focus of early Islam before the sanctuary was relocated to the Hijaz"
principal_evidence = "Early-mosque qibla orientations toward Petra (1–c.100 AH); absence of Mecca from Ptolemy's Geography; al-Azraqī's irregular Ka'ba measurements; Nabataean-Aramaic substrate of Qur'anic Arabic"
principal_critic = "David A. King — the qibla orientations reflect non-geographical 'sacred-direction' (folk-astronomical) alignment toward the Kaʿba, not aiming at Petra"
framework_reading = "Engaged, not adopted: if sound, relocates the geography of the Hanafiyya (religion of Abraham) to the Nabataean capital; converges with the corpus reading of early Islam as an Abrahamic-restoration movement in the Age of Pisces"
distinguished_from = "The Hijazi Mecca of mainstream Islamic tradition; the maximalist Hagarism revisionism of Crone and Cook; King's folk-astronomical qibla model"
+++

**Petra** (Greek <span class="arabic">Πέτρα</span> *Pétra*, "rock"; Nabataean Aramaic <span class="arabic">רקמו</span> *Raqmū*; Hebrew <span class="arabic">רֶקֶם</span> *Reqem*; Arabic <span class="arabic">ٱلْبَتْرَاء</span> *al-Batrāʾ*) is the rock-cut capital of the **Nabataean** kingdom in present-day southern Jordan. In the Wheel of Heaven framework it is engaged in two distinct registers that must be kept separate. The first is the mainstream-uncontroversial city: the Hellenistic-Roman caravan capital of the Nabataeans, annexed by Rome in 106 CE, abandoned as a major centre in late antiquity, and rediscovered for European scholarship by Johann Ludwig Burckhardt in 1812. The second is Petra as the geographic centre of a **revisionist hypothesis** — associated principally with the Canadian researcher **Dan Gibson** — that holds Petra, rather than the Hijazi city now called **Mecca**, to have been the original holy city, *qibla* (direction of prayer), and pilgrimage-focus of early Islam.

The two registers carry very different epistemic weight, and this entry marks the boundary throughout. The Nabataean-city material is `direct`: textually and archaeologically attested, uncontroversial in mainstream scholarship. The original-Mecca material is `speculative`: a contested archaeological argument that the mainstream historian of Islamic astronomy **David A. King** rejects, that is not part of the Raëlian canon, and that the corpus engages as an interpretive hypothesis rather than adopts as fact. The entry's overall `claim_type` is set to `speculative` because the page exists principally to articulate the second register; the [Muhammad](../muhammad/) entry's framework reading, which places Muhammad's birth and first revelation in the Hijaz, is left standing alongside it rather than overwritten.

## Etymology and names

The city carried different names in different languages, and the name-history is itself one strand of Gibson's argument.

### Greek Πέτρα and the modern name

The familiar name **Petra** is the Greek <span class="arabic">πέτρα</span> (*pétra*, "rock, crag"), a descriptive Hellenistic-Roman designation for the rock-cut city in its sandstone gorges. It is not the name its Nabataean inhabitants used for themselves, and it is not the name preserved in Arabic or Islamic sources. The modern global use of "Petra" descends from the rediscovery period of the early nineteenth century. Gibson stresses this point: the absence of the *word* "Petra" from the Quran or the early Islamic record says nothing about the city, because "Petra" was never the city's local name.

### Nabataean רקמו Raqmū / Hebrew רֶקֶם Reqem

The Nabataeans' own name for their capital, preserved in inscriptions, was <span class="arabic">רקמו</span> (*Raqmū*, also transliterated *Raqēmō*), corresponding to the Hebrew <span class="arabic">רֶקֶם</span> (*Reqem*) and the Greek *Rekem* known from Josephus, who reports that the city was called *Rekeme* by the local population and that the Nabataean eponymous figure of that name was buried there. Eusebius and the Madaba Map tradition likewise preserve *Rekem* / *Reqem* as the indigenous name. Gibson connects this name to the Quranic *al-Raqīm* (Quran 18:9, in the Companions-of-the-Cave passage), one of several proposed textual footholds for an early-Islamic memory of the city.

### Biblical Sela

The Hebrew Bible refers to an Edomite rock-stronghold as <span class="arabic">סֶלַע</span> (*Selaʿ*, "the rock" — 2 Kings 14:7; Isaiah 16:1), conventionally identified by many scholars with the Petra region. *Selaʿ* and Greek *Pétra* are direct translation-equivalents ("the rock"), and the identification places a fortified rock-city in this region centuries before the Nabataean florescence.

### "Mother of the Colonies"

Among the titles attested for Petra in the Roman period — preserved in the Petra papyri and inscriptions in the form *Mētrokōlonía* / "Mother of the Colonies" (the Greek <span class="arabic">μητρόπολις</span>-type honorific) — is one Gibson reads as cognate with a Quranic epithet. Quran 6:92 and 42:7 describe the sanctuary-city as <span class="arabic">أُمّ ٱلْقُرَىٰ</span> (*Umm al-Qurā*, "the Mother of Settlements / Mother of Cities"). The argument runs that the title "Mother of the Colonies" and the Quranic "Mother of Settlements" are the same honorific attached to the same city, and that the title's natural sense — a metropolis that founded daughter-colonies — fits the Nabataean trade capital better than it fits the Hijaz. This is an `inferred`-to-`speculative` reading, not a settled identification.

## The Nabataean city (mainstream account)

Petra was the capital of the **Nabataean kingdom**, an Arab people who from roughly the fourth century BCE controlled the incense and spice caravan routes between southern Arabia, the Levant, and the Mediterranean. The Nabataeans carved monumental tomb-façades, temples, and a hydraulic system of dams, cisterns, and channels into the sandstone of Wadi Musa, producing the rock-cut architecture — the Khazneh ("Treasury"), the Monastery (ad-Dayr), the Great Temple, the Qasr al-Bint — for which the site is famous. They wrote in a cursive form of **Aramaic** (the Nabataean script) that is the direct ancestor of the Arabic script, a point developed in the [Nabataeans](../nabataeans/) entry and central to the linguistic strand of the Petra hypothesis.

The kingdom reached its height under Aretas IV (9 BCE – 40 CE). In 106 CE the emperor Trajan annexed the Nabataean realm as **Provincia Arabia**, with Petra holding metropolitan rank. The city retained a substantial population through the Roman and early Byzantine periods — the [Petra papyri](../nabataeans/), a sixth-century CE archive recovered from a burned Byzantine church at the site, document a functioning Christian-era community — but a series of earthquakes (notably 363 CE) and the shifting of trade routes reduced it to a marginal settlement well before the Islamic conquests. By the time of the European rediscovery in 1812 it was known principally to local Bedouin.

This mainstream account is not in dispute. The dispute concerns what, if anything, Petra had to do with the origins of Islam.

## The Petra hypothesis (Gibson's argument)

The hypothesis that Petra was the original holy city of Islam was developed by **Dan Gibson** across *Qur'anic Geography* (2011) and, in its archaeologically most developed form, *Let the Stones Speak: Archaeology Challenges Islam* (2023), the latter incorporating statistical work by Walter R. Schumm and Zvi Goldstein and historical chapters by Chad Doell. The argument is cumulative; no single strand is decisive, and Gibson presents it as a convergence of independent lines. The corpus engages it as such — a hypothesis whose strength lies in convergence and whose individual links are each contestable.

### The qibla evidence

The load-bearing argument is archaeological. Gibson surveyed the orientation of the earliest datable mosques — the direction their *qibla* walls face — and reports that for roughly the first century of Islamic building, a large share of them do not face the Hijazi Mecca. On his classification the early mosques fall into groups he labels **Petra-facing**, **Between** (facing a point between Petra and Mecca), **Parallel** (oriented parallel to a Petra–Mecca line), and **Mecca-facing**, with the Petra orientation dominant from the outset of Islam until roughly 132 AH and the Mecca orientation emerging only later. Gibson sets a tolerance of ten degrees; mosques outside that band of any target he classifies as "unknown" rather than forcing them to fit.

He reads the chronological pattern as a migration: the earliest community prayed toward Petra; a confused "Between" period (which he associates with the governor **al-Ḥajjāj ibn Yūsuf**, c. 87–88 AH) followed; and the Hijazi Mecca orientation consolidated only afterward. The al-Aqsa mosque in Jerusalem is offered as a visible survival of the shift — its qibla wall, on this reading, faces Petra while worshippers today orient their prayer-rows off-axis toward Mecca.

The qibla data were subjected to statistical analysis by **Walter R. Schumm and Zvi Goldstein**, whose peer-reviewed work concluded that, under Gibson's assumptions, the early mosque qiblas are consistent with intentional aiming at Petra to within a few degrees, with error growing — as expected — with distance from the target. The statistical strand is what moved the argument from anecdote toward testable claim, though it tests Gibson's data and assumptions rather than independently establishing the historical conclusion.

### The problem of Mecca's antiquity

A second strand addresses the Hijazi city directly: if Mecca was an ancient sanctuary, where is it in the pre-Islamic record? Gibson and the historian **Ian D. Morris**, whose survey of the question he draws on, argue that there is no secure attestation of Mecca as a city of antiquity before Islam. The often-cited identification of Ptolemy's **Macoraba** (<span class="arabic">Μακοράβα</span>, *Geography* 6.7) with Mecca is, on this reading, a chain of assumption rather than evidence — Macoraba does not plot to Mecca's location once Ptolemy's systematic distortions are corrected, and the equation rests largely on the superficial *M-…-r-b* resemblance. Gibson's reconstruction of Ptolemy's Arabian coordinates, using rivers and known sites as control points, places Macoraba elsewhere and finds no Ptolemaic city at the coordinates of Mecca.

### The Ka'ba measurements

A third strand concerns the Kaʿba itself. The early Meccan historian **al-Azraqī** (d. c. 837 CE) recorded the dimensions of the Kaʿba as rebuilt in Muhammad's lifetime, and the four sides he gives are of *unequal* length — an irregular quadrilateral rather than a true cube. Gibson reports that an off-centre altar standing before the **Qasr al-Bint** in Petra matches al-Azraqī's four irregular measurements, including its steps, and that removing the steps yields dimensions matching the present Meccan Kaʿba. He treats this as a candidate identification of the original structure rather than a demonstrated one, noting that the Petra structure has not been excavated or studied for this purpose.

### Name migration and the two Meccas

Gibson's synthesis is that the sanctuary's furniture and names were physically and onomastically relocated. On his reconstruction the **Black Stone** was moved south during the second civil war (c. 65–70 AH) to keep it from the Umayyad armies besieging **ʿAbd Allāh ibn al-Zubayr** at the holy city — a siege Gibson locates at Petra, citing the catapult-stone deposits and barricaded structures excavated at the Great Temple — and the name "Mecca," the pilgrimage rites, and the well-name *Zamzam* followed the Stone to a new foundation in the Hijaz. For a period, on this reading, there were *two* Meccas, and later Islamic geographers such as Yāqūt, working centuries after the fact, fixed the holy places at the surviving southern site. The governor al-Ḥajjāj is cast as the pivotal "reformer" who consolidated the relocation and, in the linguistic strand, re-pointed the Quranic text.

## Connection to the Hanafiyya and the religion of Abraham

The Petra hypothesis matters to the Wheel of Heaven corpus chiefly through what it implies about the *content* of early Islam, not merely its map coordinates. Gibson's ninth chapter argues that the earliest community understood itself as seeking the <span class="arabic">حَنِيفِيَّة</span> (*Ḥanīfiyyah*) — "the religion of Abraham," a turning from polytheism to primordial monotheism — and that the word *Islām* ("submission") is a later designation, much as "the Way" preceded "Christianity." Ibn Isḥāq's account of the four pre-Islamic seekers of the Hanafiyya, set explicitly in the context of the holy city, the pilgrimage, and the Kaʿba, anchors this reading. In Gibson's geography that holy city, with its altar "of Abraham" at the city centre, is Petra.

This is the strand the corpus finds most generative, because the Wheel of Heaven reading of [Abraham](../abraham/) already treats him as a Council-recruited figure around whom a diminished lineage is *reorganized*, and already states that the Abrahamic lineage continues to produce "alliance-cultivated traditions across subsequent ages." A movement that explicitly framed itself as the *recovery* of the religion of Abraham — through the [Hagar](../abraham/)–Ishmael line specifically — is, on the corpus reading, one more cycle of that restoration pattern. See the [Hanafiyya](../hanafiyya/) entry for the full development.

## In the Wheel of Heaven framework

The framework's position on the Petra hypothesis is **engaged, not adopted**. The corpus does not assert that Petra was the original Mecca; it treats Gibson's argument as a serious revisionist reading whose truth-value is unsettled, and it declines to rewrite the existing canon entries — [Muhammad](../muhammad/), with its Hijazi birth and Cave of Hira, and [Abraham](../abraham/) — around it. What the framework does is register the *implication*: if Gibson is right, the geography of the Hanafiyya restoration sits at the Nabataean capital, and several features the corpus already tracks gain a sharper setting.

- **The Abrahamic-restoration reading.** The corpus reads the Abrahamic covenant as a recovery program and the post-Abrahamic traditions as successive "alliance-cultivated" developments. Early Islam's self-understanding as the rediscovered religion of Abraham fits this pattern directly, and a Petra setting — with a central altar attributed to Abraham and Ishmael — would give the pattern a concrete locus.
- **The Age of Pisces context.** Muhammad's career falls in the early-to-middle [Age of Pisces](../timeline/age-of-pisces/), which the corpus reads as the period of broad cultural transmission of the operational message across civilisational regions. The relocation and reformulation of a sanctuary tradition across the first Islamic century is consistent with the corpus's broader picture of traditions being edited and consolidated in transmission.
- **The editorial-suppression theme.** Gibson's reading of al-Ḥajjāj as a reformer who altered the qibla, re-pointed the text, and consolidated the relocation parallels the corpus's recurring attention to the editorial layering of received traditions — the gap between an originating contact-or-restoration event and the canonical record that later fixes it.
- **Idolatry, not polytheism.** Petra was a city of stone-cult: the Nabataeans worshipped their deities in cubic standing stones (*baetyls*), and the [Hanafiyya](../hanafiyya/) tradition sets the seekers' rejection of "the stone they went round" against exactly this idolatry. The corpus reads that contrast not as monotheism defeating polytheism — the [Elohim](../elohim/) are themselves a [plurality](../plurality-of-gods/) of embodied creators — but as a turn from worshipping dead images back to knowing the real creators, whom the [Raëlian canon](../../library/the-book-which-tells-the-truth/) says humanity had "forgotten" after the post-Sodom collapse. A Petra setting makes the idolatry literal: standing stones in the very city where the religion of Abraham was being recovered. See [Hanafiyya](../hanafiyya/) for the full development.

The framework holds all of this at arm's length deliberately. The Petra hypothesis is contested revisionism; its value to the corpus is as a hypothesis that, *if* it holds, deepens an existing reading, and the entry is labelled `speculative` to keep that honest.

## Critical reception

The principal scholarly opponent is **David A. King**, historian of medieval Islamic astronomy, who agrees that early mosques do not face the Hijazi Mecca in the strict geographical sense but rejects Gibson's explanation. King's counter-model is that early Muslims used **folk-astronomical "sacred geography"**: lacking spherical trigonometry, they oriented mosques toward the Kaʿba using non-mathematical schemes — alignment with the rising or setting of particular stars, with the cardinal directions, or with the winds associated with the Kaʿba's own walls in later astronomical-geographical texts. On King's account the early qiblas reflect these many local schemes, not aiming at Petra, and the apparent Petra focus is an artefact of Gibson's method. King's critique runs to a monograph-length treatment; Gibson's *Let the Stones Speak* devotes a chapter (and Schumm and Goldstein an appendix) to answering it, arguing that King's model multiplies alignment-schemes until it can fit any mosque and implies that Muslims spent three centuries unable to face their own sanctuary.

Beyond the King exchange, the hypothesis sits outside mainstream Islamic-studies consensus, which continues to locate Islam's origins in the Hijaz. It is distinct from — and more specific than — the maximalist source-skepticism of Patricia Crone and Michael Cook's *Hagarism* (1977), though it shares that tradition's willingness to read the archaeological and documentary record against the later Islamic literary sources. The corpus presents King's objection in its own voice precisely because the editing standards require contested claims to be held with their best counter-argument attached.

## See also

- [Nabataeans](../nabataeans/)
- [Hanafiyya](../hanafiyya/)
- [Muhammad](../muhammad/)
- [Abraham](../abraham/) / [Ibrahim](../ibrahim/)
- [Mecca](../mecca/)
- [The Qur'an](../quran/)
- [Age of Pisces](../timeline/age-of-pisces/)
- [Hebrew Bible](../hebrew-bible/)
- [Genesis](../genesis/)
- [Elohim](../elohim/)
- [List of prophets and religions](../list-of-prophets-and-religions/)

## References

### Principal hypothesis source

Gibson, Dan. *Let the Stones Speak: Archaeology Challenges Islam*. CanBooks / Independent Scholars Press, 2023. ISBN 978-1-927581-21-6. With contributions by Walter R. Schumm, Zvi Goldstein, and Chad Doell. The principal statement of the qibla, Mecca-antiquity, Ka'ba-measurement, and name-migration arguments. The authors distribute the work as a free PDF.

Gibson, Dan. *Qur'anic Geography: A Survey and Evaluation of the Geographical References in the Qur'ān with Suggested Solutions for Various Problems and Issues*. Independent Scholars Press, 2011.

Gibson, Dan. *Early Islamic Qiblas: A Survey of Mosques Built Between 1 AH/622 C.E. and 263 AH/876 C.E.* Independent Scholars Press, 2017. The underlying mosque-orientation dataset, also published as the online Qibla Tool at nabataea.net.

### Statistical engagement

Schumm, Walter R., and Zvi Goldstein. "How Accurately Could Early (622–900 C.E.) Muslims Determine the Direction of Prayers (Qibla)?" *Religions* (peer-reviewed), and the follow-up statistical appendices reproduced in *Let the Stones Speak*. Conclude that, under Gibson's assumptions, the early qibla data are statistically consistent with intentional Petra orientation.

### Critical engagement

King, David A. "From Petra back to Makka — From 'Pibla' back to Qibla." And King's monograph-length treatment comparing his folk-astronomical "sacred geography" model with Gibson's geographical-qibla theory. The principal scholarly rejection of the hypothesis; argues that early mosque orientations reflect non-mathematical astronomical alignment toward the Kaʿba rather than aiming at Petra.

Morris, Ian D. "Mecca and Macoraba." *Al-ʿUṣūr al-Wusṭā* 26 (2018): 1–60. Survey of the documentary history behind the Macoraba–Mecca identification; concludes there is no secure evidence for Mecca's antiquity in the pre-Islamic record.

### Nabataean and primary sources

al-Azraqī, Abū al-Walīd Muḥammad ibn ʿAbd Allāh. *Akhbār Makka* ("Chronicles of Mecca"). 9th century CE. Source for the irregular four-sided Ka'ba measurements.

al-Ṭabarī, Muḥammad ibn Jarīr. *Tārīkh al-Rusul wa-l-Mulūk*, vols. XXI–XXIII (the Marwanid civil-war volumes), trans. SUNY Press. Source for the Ibn al-Zubayr siege, al-Ḥajjāj, and the Ka'ba rebuilding.

Josephus, Flavius. *Jewish Antiquities* IV.7.1 (the Rekeme name for Petra). Eusebius, *Onomasticon* (Reqem). Ptolemy, *Geography* 6.7 (Arabian coordinates; Macoraba).

### Comparative revisionist context

Crone, Patricia, and Michael Cook. *Hagarism: The Making of the Islamic World*. Cambridge University Press, 1977. The maximalist source-critical revisionism against which Gibson's more specific archaeological argument is distinguished.

### Web resources

Gibson, Dan. *The Qibla Tool / Mosque Database*. <https://nabataea.net>. The searchable dataset of surveyed early mosques underlying the archaeological argument.

"Petra." *Wikipedia*. <https://en.wikipedia.org/wiki/Petra>

"Dan Gibson (writer)." *Wikipedia*. <https://en.wikipedia.org/wiki/Dan_Gibson_(writer)>