New inscriptions from Saudi Arabia and the extent of Roman rule along the Red Sea

An old blogpost from me. Originally posted in December 2014 on Since I don't want it to get lost once the old page will shut down and since it is still relevant for my Imperium Romanum map, I decided to repost it here.  

This time I want to talk about “AUSSERHALB DES REICHES? Zu neuen römischen Inschriften aus Saudi Arabien und zur Ausdehnung der römischen Herrschaft am Roten Meer” (Beyond the Empire? New inscriptions from Saudi Arabia and the extent of Roman rule along the Red Sea), by Michael A. Speidel. Since the article is in German I will write my blog post in English. The paper was originally published in the journal Zeitschrift für Papyrologie und Epigraphie (ZPE 163 (2007) 296–306). The current version is included in his book Heer und Herrschaft im Römischen Reich der Hohen Kaiserzeit, a collection of studies from 2009 and is available online from the authors page at

For decades, the archeology of Saudi Arabia was neglected. Consequently, new finds easily have the potential to literally redraw the map of the region even for otherwise quite well known epochs of ancient history. The article features such a case. In the initial part, three Latin inscriptions first published in 2004 are discussed, followed by an evaluation of their implications for our understanding of Roman influence and politics in Arabia Felix.

The first inscription was discovered in 2003 at ancient Hegra in Hedjaz, an oasis city on the Incense Road. Today it is known as Al-Hijr (Mada'in Salih) and became an UNESCO world heritage site in 2008. I had the opportunity to see the original 2012 in Berlin's Pergamon Museum as part of the Roads of Arabia exhibition and it was one of the main reasons for me to make my own up to date map of the Roman Empire.

Hegra was the major center in the south of the Nabataean kingdom that in the 1st century CE also controlled other oasis towns, such as nearby Taima or Dumatha. The kingdom was one of Rome's client states along its eastern border. When the last Nabataean king died in 106 CE, Trajan had already prepared the orders for imperial troops in neighboring provinces to swiftly move in and occupy his territory before any resistance could possibly be organized. The newly annexed state was subsequently transformed into the province Arabia. Mostly due to the remoteness of the oasis towns, most scholars believed that only the urbanized core region of the old kingdom east of the Jordan, together with the the Sinai peninsula, constituted the new province's territory and that the newly build Via Nova Traiana from Bostra to Aila, which connected a chain of forts, marked its eastern border.

In that way it was usually displayed in maps of the Roman Empire in the last decades, despite several inscriptions from the region hinting that this might not be the case. These include material from Roman soldiers found in Dumatha (Dumat Al-Jandal), Hegra, Dedan (al-'Ula) or other places and especially a bilingual inscriptions recording the involvement of an imperial governor of Arabia to settle disputes between members of the Thamudi confederation and in constructing a temple dedicated to Lucius Verus and Marcus Aurelius in Ruwwafah.  

The new inscription (AE 2004, 1620) can be dated to 175-177 CE. It commemorates the reconstruction of Hegra's defenses under the overall responsibility of the imperial governor of Arabia, Julius Firmanus, done by the Civitas Hegrenorum, supported by a detachment of Legio III Cyrenaica. Both parties are represented by two of the legion's centurions and the first citizen of Hegra, Amrus Haianis. Thus it proves that Hegra, and thus also the remaining Nabataean territories, had been fully integrated into the province, settling the dispute.

More amazing is the find of two Latin inscriptions on Farasan, the main island of the equally named archipelago, which is located more than 1000 km south of Berenice, the last Roman controlled port in Egypt. In the Latin text, the name is spelled Ferresan. 

The first inscription from Farasan is nearly complete but some parts can only be read with difficulty due to erosion. Curently, at least three readings are quoted in literature, which differ mainly in the prefect's name and the structure built.  The inscription dates to the year 144 CE and, just as in Hegra, commemorates the completion of construction work, this time a fort, done by a vexillation of Legio II Traiana Fortis and auxiliaries under the overall command of Avitus, prefect of the port of Ferresan (Portus Ferresanus) and the Herculian sea.

The toponym Pontus Herculis is otherwise not attested, but seemingly refers to the southern portion of the Red Sea leading to the strait of Bab-el-Mandeb, just as the Pillars of Hercules marked the entrance to the Strait of Gibraltar.

Although there are some hints in ancient sources which are discussed below, nobody expected to find a prefecture and a vexillation of Legio II Traina that far to the south, more than 2000 km away from their headquarter near Alexandria. The authors of the inscription's original publication deduced that the Farrasan islands were not part of the Empire. This is a possibility, since the Romans commonly maintained garrisons in their client states. However, Speidel argues convincingly that, since all other known prefectures with similar purpose were fully integrated part of their provinces, this one has to be too.

From the second inscription, only a fragment of a tabula ansata with the letters VI FERR - PR RP in two lines was found. The author interprets them as Legio VI FERRata and PRo PRaetore. Legio VI Ferrata belonged to the initial garrison of Arabia and was withdrawn during Hadrian's reign, while pro praetore also implicates a senatorial governor found in Arabia but not in Aegyptus, since that was ruled by an equestrian.

If this reading is correct, the inscription indicates that the Farrasan island were already occupied by a Roman garrison during the reign of Trajan and at this time were attached to the province of Arabia, before being transferred to Aegyptus some time before 144 CE.

In the third and largest part of his article, Speidel looks at the available evidence for Roman – southern Arabian relations and how it can help to understand the existence of the Portus Ferresanus garrison, or how its interpretation has to be changed itself in light of this new discovery. Was its sole purpose to provide a base 1000 km away from the Empire for fighting the piracy threatening the Roman merchantmen? It was apparently manned for at least several decades and the sheer existence of a distinct prefecture excludes that Portus Ferresanus was just established as a temporary staging area for a single campaign. In the next paragraphs, I will give a short chronological summary.

During the peak of their power, the Ptolemaic dynasty of Egypt took great investments to exploit the resources of the eastern desert and the coastal lands to the south. They founded a number of outposts and towns along the Trog(l)odyte coast and still controlled at least some of them well into the first century BC.

In 25 BCE, just five years after the their kingdom was annexed by Augustus, a Roman army under the command of Aelius Gallus, the prefect of Egypt, including among others a large contingent from their Nabataean clients, was dispatched to southern Arabia and advanced as far inland as Saba's capital Marib. This campaign is rather well-attested in numerous sources and generally considered a failure, since the Romans had to withdraw after loosing many soldiers to hunger and illness. Nevertheless, it marks the beginning of great and lasting changes to the political landscape of the region. In the next decades, the maritime traffic through the Red Sea massively increased, especially on the sea route between Roman controlled ports on Egypt's eastern shore and India, utilizing the Hippalos passage with the monsoon winds. This lead to an decline of the old caravan kingdoms along the incense route, while Himyar and Hadramawt, both with coastal access, became the dominant powers of the region. Himyar with its capital Sapphar was centered in the fertile highlands of modern Yemen, while Hadramawt, to the east, mostly controlled the frankincense production. For most of the first and second century CE, Himyar controlled Saba and according to the mid first century Periplus of the Erythraean Sea, at least for some time also the ports of Azania on the opposing coast in modern Somalia.

In modern Baraqish (ancient Athr(o)ula or Sabaean Yathill), the fragment of a bilingual funeral inscription for the Roman Eques P. Cornelius in Latin and Greek was found. The term Eques can designate an ordinary cavalryman or a member of the equestrian order. So far it was attributed to Aelius Gallus campaign, who established a base there, but it can not be dated precisely. In light of the Farasan discovery, it could also be from a later date, especially since it would be highly unusual to erect a funeral monument in hostile territory.

About 6 BCE, bilingual descriptions from Sirwah attest the presence of Nabataean community in the heartland of Saba and on the route between the capitals Marib and Sapphar, distant from the old Incense Road.

A siege of Najran by Nabataean forces is attested, as well as finds of their ceramics on the Farasan islands, which can possibly be identified with the islands of Malichos mentioned by Pliny (NH 6.175) and Ptolemy (6.7.44). This allows for the hypothesis that the Roman outpost did have an older Nabataean predecessor, which is given some more weight if Portus Ferresanus was indeed at first part of Roman Arabia.

Between the end of the 1st century BCE and the middle of the first century CE, Saba emits a series of coins portraying Emperor Augustus. Slightly later, the Periplus describes Charibael, king of Himyar and Saba as friend of the emperors. If correctly interpreted, a roughly contemporary Chinese source called Weilue (Peoples of the west) describes the lands around the horn of Africa as subordinate to the Roman Emperor.

As a preparation for his offensive into south Arabia, Aelius Gallus had ordered the construction of a naval fleet, whose ultimate destiny remains uncertain. However, ostraca and papyri from Egypt's Red Sea ports attest the continued presence of Imperial navy vessels throughout the first century.

In his work about the first Jewish War, Josephus Flavius included a speech he attributes to the Roman client king Agrippa II., in which he describes the extent of the Roman Empire (BJ 2.16). Here, the province of Egypt is described as extending as far as the Ethiopians and Arabia Felix.

Also writing during the Flavian dynasty, Pliny the Elder remarks that only Aelius Gallus carried Roman arms into Arabia Felix (NH 6.160), excluding other large scale expeditions to the region until his time.

Writing during the reign of Trajan, the historian Tacitus describes Germanicus' tour of Egypt in 19 CE (Ann. 2.61). Elephantine and neighboring Syene are mentioned as the former border of the Roman Empire, which now (in his days) extended to the Red Sea. Also under the reign of Trajan, the old canal between the Nile and the Gulf of Suez, which had become silted since the time of Ptolemaios II, was reconstructed. The fourth century historian Eutropius mentions in his Breviarum (8.3) that Trajan created a fleet in the Red Sea.

Both measures, as well as the establishment of Portus Ferresanus are in line with Trajan's expansionist policy and perhaps were part of his grand scheme to conquer the Parthian Empire and rival Alexander the Great. By annexing former client states, a process already started by Vespasian in the east, the Empire also had to take over their former responsibilities, like the protection of borders and trade routes. 

In his Roman oration (70.), Aelius Aristides gives a cryptic list of wars fought at the edges of the Empire during his time, including one that made an end to the “misery of the peoples at the Red Sea who are unable to enjoy the blessing of the present”, thus pointing to another Roman military intervention during the existence of their outpost on Farasan.

And finally there is the base of an statue with a Greek inscription which was found in Sapphar that once apparently carried the statue of Roman Emperor. It can be roughly dated to the 2nd or 3rd century.

Speidel concludes that, after the annexation of Egypt, Rome tried to gain control over all former Ptolemaic possessions and to establish itself as hegemonic power in the border zone, just as it happened with Nabataean kingdom 136 years later. The campaign of Aelius Gallus has to be seen in this context. The Roman Empire considered the states around the Red Sea as being in its sphere of influence and tried to keep them in a pro-Roman stance. This required the credible capability to enforce the Emperor's will upon his “friends” if necessary, for which the base on Farasan would be highly useful.

Speidel further explores this topic in his forthcomming article Fernhandel und Freundschaft. Zu Formen römischer Wirtschaftsförderung am Roten Meer und am Indischen Ozean (Long distance trade and friendship. Ways of Roman support for economic development at the Red Sea and Indian Ocean). The manuscript is again available from

Despite the decline of Red Sea trade during the third century crisis that would never again reach the same scale as during the Principate era, Roman interest in southern Arabian affairs did not cease. As late as the 6th century, the allied Christian Eastern Roman and Aksumite Empires were engaged in proxy wars with non-Christian, pro-Persian factions over the control of Himyar, which by this time incorporated all of southern Arabia. Only the Persian conquest of Himyar in the 570ties, which was soon followed by the rise of Islam, finally brought Roman influence to an end.

Write a comment

Comments: 1
  • #1

    Jakub Jurasz (Monday, 30 August 2021 11:59)

    Nice post. I found it on google whilst I was reading the Fate of Rome by Kyle Harper and I was curious where the Farasan Islands are. Truly impressive.