Preface: Punt is thought to have encompassed areas of Sudan, Eritrea, Northern Ethiopia and parts of Northern Somalia, reaching down to the region of Puntland. It was also a society of diverse ethnic groups. While this article focuses on Punt within Eritrea, it does not dismiss the likelihood that Punt also included regions of Modern-day Somalia, Djibouti or Sudan.

Date Created: 4/04/2026

Last Updated: 11/04/2026

The article is still in its draft stages & will be edited further…

I will begin by introducing Punt, then examine the evidence suggesting that Eritrea was likely an integral part of Punt, and finally provide an overview of its history.

Introduction

The ancient civilisation known as Punt has long perplexed historians and archaeologists due to the limited information available about it. Punt is mentioned in Egyptian sources from approximately 2500 BC to 1000 BC, a timeframe that spans the Old, Middle, and New Kingdom periods of ancient Egypt. This shows a long-standing and sustained relationship between Punt and Egypt for over a thousand years.

In ancient Egyptian, Punt is spelled pwnt, written with the hieroglyphs as1:

However, despite the numerous expeditions recorded by the Egyptians to this mysterious land, the precise location of Punt remains a subject of ongoing scholarly debate.

Punt was renowned in ancient Egypt for its production of Frankincense (Ancient Egyptian: snTr2) and Myrrh (Ancient Egyptian: antyw3), earning it the titles “The Land of Plenty”4 and “God’s Land” (Ta-Netjer5) due to its vast resources.

What is Myrhh and Frankincense & Why was it important?

Both myrrh and frankincense are obtained by cutting & peeling 6 resin-producing trees: myrrh from Commiphora and frankincense from Boswellia (with Pistacia sometimes associated with both). These trees are found across the Mediterranean, Syro-Palestine, East Africa, and Arabia7.

Over time, a substance oozes from these cuts and hardens upon contact with the air, forming lumps of varying sizes, often called “tears”. These are then collected and used for various purposes, such as Incense, where they are burned to produce fragrant smoke8.

In ancient Egypt, incense was regarded as inherently divine. Even the word for frankincense (snṯr) could, in certain contexts, mean “to make divine”9. As a result, incense was used to cleanse and prepare spaces for sacred ritual activities.

Interestingly, Punt was never invaded by Ancient Egyptian military forces, nor did the Egyptians establish any permanent settlements there10. This is despite the fact that ancient Egyptians may have spent several months in Punt, as they had to wait for favourable sailing conditions before making a safe return journey to Egypt11.

Trade goods from Punt are often described as “marvels” (“biAt”). Scholars interpret this as a distinct term referring to trade or gift exchange, rather than tribute, and note that it was used for regions beyond the political and military reach of the Ancient Egyptians12.

The Egyptians probably found it more convenient to depend on the Puntites for the production and supply of rare trade items, instead of trying to conquer and transform Punt into a subject territory. The challenges of navigating the distant location of Punt, whether overland or through the red-sea probably rendered the idea of a military expedition impractical.

Secondly, the Egyptians may have held the Puntites in very high esteem. Even some of their gods, such as Hathor (the goddess of love, fertility, and more13), were closely associated with Punt. Hathor was one of the earliest Egyptian deities, dating back to the predynastic period, and, most importantly in this context, she was frequently described as originating from Punt14.

Where was Punt located?

The precise location of Punt remains a topic of considerable debate, with theories suggesting various regions, including the southern Arabian coast, Eritrea, Sudan & Somalia. Currently, a majority of scholars lean towards the view that Punt likely included the territories of present-day Somalia, Djibouti, and Eritrea15. The lack of locational details in ancient writings regarding Punt, along with the sparse archaeological findings make it difficult to pinpoint the location of Punt.

An inscription from the 26th Dynasty (~672–525 BC16) records an interesting event described as a “great marvel”, in which “the heavens rained upon the mountains of Punt”, causing the Nile to flood17. This account lends further support to the view that much of Punt lay in regions corresponding to modern Sudan, Eritrea, and Ethiopia, near the sources of the White and Blue Nile18.

In the next section of this article, I’ll provide evidence showing why Eritrea was a core component of Punt.

Subscribe now

Connection to the Eritrean region

Bia-Punt

What is Bia-Punt?

Punt often acted as an intermediary for other African nations in their trade with Egypt. While little is known about most of these "other" nations, records and research have shed some light on three: Amu, Irem, and Bia-Punt. Among these, Bia-Punt is the most documented and is thought to be a region within Punt19, most likely located in Northern Eritrea. This area was particularly noted for its mining activities, suggesting that Bia-Punt was a significant source of minerals and possibly other valuable resources in the trade networks between Egypt and the African interior.

In ancient Egyptian, Bia-Punt is written bjꜣ pwnt, meaning ‘mines of Punt’20.

Sesostris I

There are several mentions of Bia Punt, for example, during the reign of Ancient Egyptian Pharaoh Sesostris I (1956-1911 BC), a shrine belonging to an official named Ankhu was discovered. The shrine consisted of three discarded stone anchors, and it contained the following inscription, mentioning the “mining region of Punt”:

The Tale Of The Shipwrecked Sailor

The Tale of the Shipwrecked Sailor is among the very few surviving fictional narratives from Egypt’s Middle Kingdom21. It recounts the journey of a sailor who, after being shipwrecked during a voyage to a mining region, encounters the prince of Punt (who is a giant serpent). The prince boasts of immense wealth, particularly myrrh for incense production, alongside oils, spices, and, most interesting of all, a range of raw animal goods, including elephant tusks and monkeys.

These same commodities (elephant tusks & monkeys) are later identified by Greek and Roman writers, millennia afterwards, as primary exports of Adulis and its surrounding region22. This lends further weight to the argument that Adulis functioned as one of the principal ports and commercial hubs of Punt.

Here are excerpts from The Tale of the Shipwrecked Sailor that relate to Punt.:

“….I was going to the mine of the king. I went down to the sea in a ship of one hundred twenty cubits in length and forty cubits in width. One hundred twenty sailors were in it of the choicest of Egypt. Whether they looked at the sky or they looked at the land, their hearts were braver than lions. They could foretell a storm before it came, foul weather before it occurred….

Then he laughed at me for what I said was foolishness to him. He said to me, “You are not rich in myrrh being an owner of incense. It is I who am the lord of Punt and the myrrh, it belongs to me. And the incense that you spoke of bringing, it is abundant on this island. When it happens that you leave this place, it will not occur that you will see this island again, it having become water”…

Then I placed myself upon my belly my arms bent in respect before him. Then he gave to me a quantity of myrrh, heknu oil, laudanum, hesayt spice, tishpes spice, perfume, eye-paint, giraffes’ tails, great lumps of incense, elephants’ tusks, greyhounds, monkeys, baboons and all kinds of precious things. Then I loaded them upon this boat. It happened as I placed myself on my belly to give thanks to him that he said to me, “Behold, you will approach home in two months. You will be full, you will embrace your children, you will be young in the home where you will be buried”. - Excerpts from The Tale of the Shipwrecked Sailor, a translation by W.K. Flinders Petrie, 1892.

Please be aware that various translations of the story exist, with each one differing slightly in their interpretation.

The Tale of the Shipwrecked Sailor mentions a serpent prince, which bears a striking resemblance to the worship of the serpent Arwe, which is attested across the Eritrean and Ethiopian highlands prior to Christianity. Interestingly, several royal chronicles begin with Arwe as the first ruling figure23. This raises the question of whether these traditions preserve a distant memory of a serpent cult that existed in the region during Puntite times.

Precambrian rocks

Another line of evidence linking Punt, specifically Bia-Punt (the mining region of Punt), to Eritrea is the region’s geology. Over 60% of Eritrea is composed of Precambrian rock formations24, which are particularly rich in valuable minerals, including gold. These mineral resources align closely with what the ancient Egyptians described as electrum.

Precambrian rocks are some of the oldest on the planet, forming the Earth’s early crust. These ancient geological formations are known for their potential to contain valuable mineral deposits. In Northeast Africa, areas with old metamorphic (Precambrian) rocks are particularly noted for their gold yielding potential. The presence of these rocks is a key indicator for geologists and mining companies when exploring for gold deposits25.

Adulis and Ancient Egyptian Baboons

Adulis, located on the coast of present-day Eritrea, was an important maritime port in the Erythrean Sea throughout antiquity. Its history can be traced back to at least the first millennium BCE26, although it’s likely much older. This port city was pivotal in facilitating the movement of ships, goods, and ideas among major ancient centres such as Egypt, Arabia, and India.

One of the exports of Adulis was monkeys and baboons, as noted earlier in this article. Writing in the 1st century AD, Pliny the Elder records that sphingiæ (possibly sphingion, a type of ape27) were brought from Adulis. However, the origins of this monkey trade are far older, as will become clear in the following section.

Hermopolis Magna

Hermopolis Magna, currently known as el-Ashmunein, is situated in Upper Egypt and was known as a significant religious centre dedicated to the worship of the god Thoth during the Pharaonic period ⁹. Thoth was venerated as the god of writing, wisdom, and the moon, and held the title of vizier to the gods. His significance in the Egyptian pantheon is reflected in the grandeur and devotion evident in the sanctuary and temple complex of Hermopolis Magna28.

Hermopolis Magna, which can be broken down to Hermo → Hermes (God of Trade, Travellers etc…) , polis → city & Magna → Great, so the literal translation is “Great city of Hermes”29. Hermes was the Ancient Greek god equivalent to Thoth30.

In Ancient Egypt the city was called Khmunu, which means “City Of Eight” which is in reference to eight gods whos centers of worship were here31.

As I noted in my article on Adulis, the 6th-century explorer Cosmas32 records that an engraving of Hermes (Mercury in Roman) appeared on the back of the Adulis Throne. This suggests an even deeper connection, showing that the significance of this deity extended from ancient Egypt all the way to Adulis.

This religious site is famed for its numerous colossal statues, each weighing approximately 35 tons. These statues represent Papio hamadryas (a species of baboon) as manifestations of Thoth33.

Connection to Adulis

Further evidence linking Adulis to ancient Egyptian baboons emerged in 2023, when a team of researchers conducted genetic analyses on baboon remains from ancient Egypt, particularly those from Gabbanat el-Qurud, also known as the “Valley of the Monkeys”. Mitochondrial genomes from a mummified baboon at Gabbanat el-Qurud, together with fourteen specimens held in other museums, were analysed and revealed a striking pattern. Genetic results link these baboons to modern populations of Papio hamadryas found in Eritrea, strengthening the case for Adulis being the major source for Egypt’s sacred baboons34.

Obsidian at Mersa/Wadi Gawasis, Linked to Eritrea

The ancient port of Mersa/Wadi Gawasis in Egypt played an important role in the ancient Red Sea trade, especially with Punt. The site has been identified as the ancient Egyptian port of Saww35.

One of the many artefacts linking this port with Punt dates to the reign of Pharaoh Amenemhat IV (c. 1786–1777 BC)36. A wooden storage box, likely used during an expedition to Punt, was discovered bearing an inscription that mentions Bia-Punt.

One of the most intriguing finds at this port was obsidian, a type of volcanic glass. Given that obsidian does not occur naturally in Egypt, its presence at the site provided clear evidence of long-distance trade37.

Linking it to Adulis

Another significant discovery at the port of Wadi Gawasis consisted of tools used by the ancient Egyptians for various activities at the port, including scraping and cleaning ship timbers as well as cutting tasks38. In particular, six such artefacts made of obsidian were found. Archaeologists conducted geochemical analysis to determine the origin of these obsidian tools.

Specifically, these obsidian tools were compared with geological samples from across Eritrea, Ethiopia, and Yemen. The results revealed that four of the six artefacts closely matched obsidian sourced from Kusrale in Eritrea39.

Kusrale is a basin in the Gulf of Zula, where obsidian cobbles and pebbles accumulate from the surrounding mountains40.

Possible Origins of Habesha & Kebessa

Before Queen Hatshepsut’s expedition to the land of Punt, she consulted an oracle of the god Amun. The oracle instructed her to undertake the journey to Punt. An inscription preserving the oracle’s message survives in both Hatshepsut’s temple at Karnak and at Deir el-Bahari41.

Interestingly, in this inscription, the term ḫbś.tıʾw is used in reference to Punt. This term means “Bearded Ones”42.

The hieroglyphics above can be further broken down into:

• The hieroglyph below represents ḫbś.tıʾw, meaning “Bearded Ones”43.

  

• The hieroglyph below represents n(.ıʾ )w tȝ-nčr44, which means “Gods Land”, which in this case refers to Punt.

Some scholars, such as Francis Breyer, have proposed that the term ḫbś.tıʾw may represent the ethnogenesis of the name Habesha45, now used as a modern pan-ethnic identifier. He argues that, during the time of Punt, the term functioned in a geopolitical sense.

The vocalisation of ḫbś.tıʾw can be khebsi46; could this be an early etymology of the term Kebessa, used to refer to the highland people of Eritrea47?

Note: Francis Breyer suggests that the “Puntites” were also referred to as “hbstiw,” which later evolved into “hbst” during the Aksumite period. In Arabic inscriptions, this term appeared as “habasat”, leading to the modern term “habesha”48.

In my opinion the highlands of Eritrea were probably one of the first places where Habeshas originated from, with the first inhabitants near the coastline near Adulis being given the name Khebsi.

Another interesting clue lies in the way the ancient Egyptians depicted the Puntites. They were typically portrayed with red or copper-toned skin, similar to the usual representation of Egyptians of themselves, which contrasts with the way Nubians were depicted in Egyptian art.

For example, in the tomb of Rekhmire (vizier under Thutmose III and Amenhotep II49), a beautiful relief depicts various foreign delegations presenting goods to the Egyptians. Specifically, in the top left, Puntites can be seen50.

In my view, Punt likely comprised a diverse range of ethnic groups, given the large geographical region it encompassed. Cushitic-speaking populations were probably the most predominant, but Nilotic and Semitic groups were probably also present. By the later phases of Punt, particularly around the time of Queen Hatshepsut (c. 1500 BC), Semitic-speaking communities may have been more common in certain areas, especially in coastal regions such as around Adulis.

Conclusion

Based on the evidence outlined above, including references to mining activity in Bia-Punt, genetic analyses of Papio Hamadryas baboons in ancient Egypt linked to modern populations in Eritrea, and the sourcing of obsidian fragments traced to Eritrea, it is reasonable to conclude that Eritrea formed at least part of the land of Punt, if not one of its main regions.

Early Punt (2500BC-2040BC)

Pharaoh Sahure (~2487BC – 2475BC)

The earliest currently known reference to Punt is dated to the reign of Pharaoh Sahure (~2487BC - 2475BC), in the passageway of his Pyramid at Abusir. This scene depicts the Pharaoh celebrating the gathering of a Frankincense tree from Punt.

The inscription accompanying it is as follows:

Although not entirely certain, the earliest reference to Punt may actually date to the reign of Pharaoh Sneferu (c. 2613–2589 BC), where “fresh incense” is recorded as being brought to Egypt, some scholars suggest this may have originated from Punt51.

The Palermo Stone (2490-2475BC)

The Palermo Stone is one of seven surviving fragments of the Royal Annals of Ancient Egypt from the Old Kingdom. It records the succession of Egyptian kings from the Predynastic period through to the Fifth Dynasty. Notably, the stone preserves one of the earliest known references to the land of Punt. The hieroglyphic inscriptions describe several goods imported from Punt during the last year of the reign of the Fifth Dynasty Pharaoh Sahura52(~ 2490–2475 BC53).

The inscription is as follows:

What is electrum? Electrum usually referred to an alloy of gold and silver, but in the context of imports from Punt, it likely referred to pure gold.54.

Green malachite refers to malachite powder, which was used by the ancient Egyptians for cosmetic purposes55.

Harkhuf (2287BC-2184BC)

Harkhuf served as part of the nobility of southern Egypt56 during the reigns of Pharaoh Merenre (c. 2287–2278 BC) and Pharaoh Pepi II (c. 2278–2184 BC)57. He led at least four expeditions southwards into Nubia/Yam. Most importantly for our purposes, his inscriptions also contain a reference to Punt.

The inscription is as follows:

Analysis

Firstly, the inscription records that a Pygmy was brought back from Punt. It is unlikely that this individual originated from Punt itself rather, Puntite traders may have obtained the person through trade with neighbouring interior regions and then exchanged them with the Ancient Egyptians. This interpretation is supported by Egyptian depictions of the Puntites, who, as we discussed earlier, are usually portrayed with physical features similar to those of the Egyptians.

Pygmies were known as dng in Ancient Egyptian and were highly prized for their rarity among the nobility, for example in one instance, their dance was even compared to that of the gods58.

This inscription raises a strong possibility that forms of human trade, including slavery, were conducted by the Puntites.

We gain insight into an even earlier voyage to Punt during the reign of Pharaoh Djedkare Isesi (~2414–2375 BC59), led by an official named Bawerded. At the ancient Red Sea port of Mersa Gawasis60, inscriptions bearing the name of the same Bawerded have been discovered, strongly suggesting that this harbour served as a main departure point for Egyptian trade missions and expeditions to Punt at least from the mid-half of the 2nd millineum BC.

During the reign of Djedkare Isesi, power gradually became decentralised, allowing regional governors (known as nomarchs) to gain greater authority. This shift would eventually lead to the start of the First Intermediate Period61.

It is also evident that the Ancient Egyptians greatly valued this dancing Pygmy from Punt, who was no doubt regarded as an exceptionally rare curiosity. The traditional dances performed by the individual must have provided considerable entertainment for the Pharaoh and his royal court. Although Harkhuf’s quotation of the king, expressing a desire to see the Pygmy more than any other gift from Punt, is probably a exaggurated.

Pharaoh Pepi II(2278–2184 BC)

During the reign of Pharaoh Pepi II(2278–2184 BC62), an official from Elephantine named Pepi-nakht records that a shipbuilder and a group of Egyptian soldiers were killed by Bedouins while constructing ships intended for an expedition to Punt63.

The inscription is as follows:

Analysis

This inscription demonstrates that the route to Punt could be highly perilous, even before the ships had set sail. It proves that the importance of trade with Punt outweighed the risks involved, including, in this instance, the loss of soldiers and shipbuilders.

The Middle Kingdom (2040–1795 BC)

Between the Old Kingdom (2686–2181 BC64) and the Middle Kingdom (2040–1795 BC65) of ancient Egypt lies the First Intermediate Period (2181–2040 BC66). During this time, power became decentralised and was exercised by powerful local governors known as nomarchs. This fragmentation reduced the state’s ability to undertake large-scale projects that required the combined resources of Egypt and the coordination of a strong central authority. In addition, documentation from this period is scarce, leaving us with limited information about external regions such as Punt67.

Starting with the ruler Intef(~2125BC68) and culminating in the reign of Pharaoh Mentuhotep II( ≈~2055BC - ~ 2004BC69), who unified Upper and Lower Egypt, the Middle Kingdom was established, and an era of centralised authority was restored. Soon after, trade with Punt was re-established.

Henu (2004–1992 BC)

During the reign of Pharaoh Mentuhotep III(2004–1992 BC70), an official named Henu, along with three thousand militia men, was tasked with constructing ships and sailing across the Red Sea to Punt. The expedition later returned with several luxurious resources from Punt to present to the Pharaoh.

The inscription reads as follows:

Historians suggest that expeditions to Punt likely departed around the 4th of September, with August to September considered the most favourable time to set sail from the Red Sea coast of Egypt, this also aligns with later roman-era primary sources71.

Analysis

Firstly, it appears that the difficulties earlier faced by the ancient Egyptians during the reign of Pharaoh Pepi II (particularly in reaching the Red Sea coast and constructing ships due to the presence of local Bedouin tribes) remained an issue even several centuries later, hence the need for Henu to be accompanied by 3,000 soldiers.

Secondly, the first stage of the journey, prior to the construction of the ships, required crossing the Eastern Desert. Throughout this part of the expedition, Henu states that he provided his men with adequate and much-needed supplies. Once they reached the coast, they built the ships and sailed to Punt with their trade goods. The expedition appears to have been successful, as they later returned with various gifts for the Pharaoh as well as stone blocks intended for the construction of royal temples.

However, an important question remains: which ports did Henu and his men use to build these ships? Fortunately, stone anchors dating to the reign of Senusret I (~1965 BC – ~1920 BC72) reveal the likely location from which many Middle Kingdom, and likely even earlier, expeditions to Punt began.

Senusret I (~1965 BC – ~1920 BC)

On the northern Red Sea coastline lies the site of Mersa/Wadi Gawasis, a port known to the ancient Egyptians as “Saww”73. The site contains numerous artefacts dating to the 12th and 13th Dynasties of ancient Egypt (~1985–1650 BC). Many of these finds, including stone anchors and wooden crates, are linked to its function as a harbour where ships were built, docked, and launched for voyages across the Red Sea, particularly to the land of Punt.

Mersa means small harbour in arabic while Gawsis is the plural form of Gasus which means a type of small boat used for observation of enemy ship74. So a literal translation could be “Port Of The Spy Boats”.

The port was primarily used during the Middle Kingdom (~2055–1650BC), however, evidence of settlements during the Old Kingdom also exists, as discussed earlier in the article75.

Specifically, at Wadi Gawasis, a stela was found attributed to the reign of Pharaoh Sesostris I(~1965 BC – ~1920 BC76), which describes a journey to the land of Punt.

Wadi Gawasis derives from the Arabic word wadi77 meaning valley, and gasus (plural gawasis), meaning spy (a term also used for small boats employed to observe enemy ships). A literal translation could therefore be “Valley of the Spy Boats”.

The inscription is as follows:

Analysis

This inscription is straightforward, it states that in the 24th year of the reign of Pharaoh Sesostris I, ships were assembled at Mersa/Wadi Gawasis and successfully sailed to the “mining region of Punt”/Bia-Punt (which we already concluded earlier in the article was likely northern Eritrea & southern Sudan) and back, returning as usual with a great quantity of trade goods.

About 200 metres west of the previously discussed Stela stands another inscription describing a journey to Bia-Punt during the reign of Pharaoh Sesostris, I led by the vizier Intefoqer78.

The inscription is as follows:

From this inscription, we learn that part of the ship construction began further inland at the town of Koptos. It also provides a rough estimate of the number of people involved. It appears that not only trade personnel and ship crews travelled to Punt, but also scribes.

This offers a new perspective on the activities of Ancient Egypt in relation to Punt, providing evidence that these expeditions were not solely focused on trade but likely also involved geographical observation and documentation of the region and its inhabitants, which would explain the presence of scribes.

Pharaoh Amenemhat II (~1920 BC – ~1878 BC)

During the 28th year of the reign of Pharaoh Amenemhat II (~1920 BC – ~1878 BC79), a stela was erected at Wadi Gawasis, which directly mentions Saww (Wadi Gawasis) as the place from which ships departed for Punt.

The Inscription is as follows:

Pharaoh Amenemhat III (~1855BC - 1808BC)

A stelae discovered at Mersa Gawasis, dating to the reign of Amenemhat III, describes expeditions to Punt and Bia-Punt. It records the journeys of two high officials, Nebsu and Senbef, who travelled to the mining regions of Punt (likely in northern Eritrea and parts of eastern Sudan).

Pharaoh Senusret III (~1874 BC - ~1855 BC)

The final known expedition of the Middle Kingdom is attributed to the 5th year of the reign of Pharaoh Senusret III (~1874 - 1855 BC80). Archaeologists discovered fragments of pottery bearing Sesostris III’s royal cartouche and inscriptions indicating that the goods were destined for Punt.

Shenu (meaning “to encircle”) was the Ancient Egyptian term for a cartouche, an oval enclosure used in inscriptions to contain a pharaoh’s name. These cartouches were significant because they symbolised protection, as a name had to be preserved for the individual to enter the afterlife81.

Pharaoh Amenemhat IV (1808BC-1799BC)

As noted earlier in the article, boxes of wooden crates dating to the reign of Pharaoh Amenemhat IV (1808–1799 BC82) bear inscriptions referring to the “marvels of Punt” (bjȝwt Pwnt), providing clear evidence that trade with Punt continued right up to the end of the Middle Kingdom83.

Second Intermediate Period (~ 1795 BC –1650 BC)

Just like the First Intermediate Period discussed earlier, the Second Intermediate Period(~1795BC – ~1650BC84) brought about a significant decline in the Egyptian state’s ability to organise expeditions to distant lands such as Punt. However, while the First Intermediate Period was largely caused by internal strife and the growing power of provincial governors, the Second Intermediate Period was different.

During this time, a foreign group known to Greek writers as the Hyksos, and referred to by the ancient Egyptians as “Asiatics”, entered Egypt from the Levant and occupied parts of northern Egypt for more than a century, disrupting the natural order.

The Hyksos were not the only foreign threat to Egyptian power. To the south, the Kingdom of Kush was rising in strength, gradually dominating intermediary trade with regions further south, including Punt85.

It wasn’t until the Kamose & his Pharaoh Ahmose I (~1550BC - ~1525BC86) that the Hyksos were driven away and Ancient Egypt was once again unified, and a new period called the New Kingdom (1550–1069 BC87) began.

Punt's Golden Age (1550BC-1350BC)

Pharaoh Hatshepsut (1472BC to 1458BC)

Pharaoh Hatshepsut (~1472 BC to ~1458 BC88) ushered in a new era of expeditions to Punt. Unlike any journey before or after, her expedition was recorded in incredible detail, providing us with one of the clearest glimpses into the Land of Punt.

She chose to construct her mortuary temple in a valley previously utilised by her ancestor, Mentuhotep II, for his own temple. However, Hatshepsut’s vision was to create something far grander. Her temple is known as Djeser-djeseru or ‘the Holiest of Holies’.

Hatshepsut means “foremost of noblewomen”89 in ancient Egyptian.

Within her temple is a section dedicated to her expedition to Punt, featuring beautifully detailed reliefs that depict the voyage, the arrival in Punt, the exchange of goods, the return journey and finally the presentation of Puntite goods to Queen Hatshepsut.

Motive

So why did Pharaoh Hatshepsut seek to travel to Punt? The motives were multifaceted. Prestige was certainly an important factor, but establishing a direct trade relationship with Punt (thereby eliminating intermediaries such as Nubia) was another key objective. She also likely aimed to introduce and cultivate some of Punt’s valuable goods, particularly incense, within Egypt itself90.

At Hatshepsut’s temple at Deir el-Bahari, a beautiful scene shows the queen seated upon her throne, addressing her royal court and declaring her decision to launch an expedition to the Land of Punt. The accompanying inscriptions present her mission to Punt as a divine command from her father, Amun-Ra.

Originally a protective deity of the pharaoh, Amun rose to prominence during the New Kingdom after Ahmose I credited his reunification of Egypt to him. Combined with the sun god Ra as Amun-Ra, he became a supreme deity, regarded as king of the gods and creator of all things, and remained central to the religious life of ancient Egyptians91.

The inscription is as follows:

The claim that Hatshepsut’s expedition was the first to reach and “tread” the land of Punt is hyperbolic, intended to enhance her prestige, as discussed earlier, such expeditions had taken place long before her reign.

Statue Of Punt?

Interestingly, an inscription records that Queen Hatshepsut ordered a granite statue depicting both Amun-Ra and herself to accompany the expedition, to be set up in Punt.

Could this have been a precursor to the Monumentum Adulitanum, the stele inscription recounting the conquests of Ptolemy III (~246AD), which was later transported and erected at Adulis?

The Journey To Punt

The journey to Punt began in the ninth year of Pharaoh Hatshepsut’s reign and was led by the high official Nehisi. The expedition likely followed earlier routes, travelling from Koptos across the Eastern Desert to the Red Sea ports of Sa’waw (Mersa Gawasis) or Quseir92.

The first scene at her temple at Deir el-Bahari depicts the Puntites setting off from Egypt (on the far left) with the inscription accompanying it reading:

To the right, the ships depict Ancient Egyptians sailing the Red Sea towards Punt, filled with crates of goods and gifts, alongside oarsmen and guides, while various sea creatures are shown beneath them. The inscription alongside it reads:

Arrival At Punt

The next scene depicts the official Nehisi, followed by an officer and eight Ancient Egyptian soldiers carrying shields and spears. In front of Nehisi are gifts for the Puntites, including five large metal bracelets, two leg bangles, an axe, a dagger, and eleven beaded necklaces93.

To their left stands Parehu, the chief of Punt, accompanied by his wife Ati, their two sons, and a daughter. Behind the royal family are several Puntites with a donkey laden with goods, further back are the trees of Punt, a cow lounging, and Puntite houses accessed by ladders.

The accompanying inscription reads:

Again, the reference to “a land that Egyptians did not know” is clearly hyperbolic. As discussed earlier, numerous expeditions had already been undertaken to punt however, it may have been several decades or even centuries since the last Egyptian expedition to Punt (see the Second Intermediate Period above), meaning it likely no longer existed within living memory of the Puntites.

The Meeting

In the scene above, the chief of Punt, Parehu, along with his wife and several Puntites, presents offerings of gold, boomerang-shaped objects, and myrrh and frankincense to the ancient Egyptian delegation led by Nehisi.

Nehisi then invites them to a feast where they enjoy the gifts brought from Egypt. The accompanying inscription reads as follows:

Homes

To the left of the aforementioned meeting is a now badly damaged scene depicting the homes of the Puntites. These huts, raised above the ground and accessed by ladders, are surrounded by incense trees, some of which have monkeys hanging upon them. Other wildlife, including cattle and a giraffe, is also shown. The scene also depicts Puntites gathering incense, with others carrying it in large bundles.

The accompanying inscription is as follows:

The Return Journey

The following scene depicts Ancient Egyptian sailors loading their ships with the goods acquired from Punt, including ivory, leopard skins, baboons, and even entire frankincense trees, as well as some children of Punt. Further to the right, several ships can already be seen setting sail.

The baboons depicted are the same Papio hamadryas discussed at the beginning of this article, and are associated with populations found across Eritrea.

The accompanying inscription reads:

Arrival In Egypt

Next, a scene depicts both Ancient Egyptians and Puntites (including four Puntite princes) presenting the goods they brought from Punt to Pharaoh Hatshepsut.

The next scene shows Pharaoh Hatshepsut offering a selection of Puntite goods to the god Amon-Ra; incense trees, rhinoceros horns, tortoise shells, and more can be seen. We can also observe many fully grown incense trees.

First Prophet Of Amun - Hepusonb

During the reign of Pharaoh Hatshepsut, the High Priest of Amun was a man named Hepusonb. In his temple, a relief depicts incense being collected from a tree in Punt, likely highlighting the importance of incense in the rituals he conducted94.

We also see a depiction of a chariot being constructed, likely used to meet the Puntites upon their arrival at the port of el-Quseir (Mersa Gawasis)95.

Thutmosis III

During the reign of Thutmosis III(~1458-1425BC96), we have several pieces of evidence of trade with Punt being conducted. For example, two inscriptions, the first dated to roughly 1447 BC & the second to 1442BC, refer to various goods such as myrrh, gold (or electrum), slaves, and different types of cattle being brought from Punt.

The first inscription is as follows:

The second inscription is as follows:

Tomb Paintings

Dating to the reign of Thutmose III, tomb paintings found in the tomb of the Second Prophet of Amun, Puyemre, depict the pharaoh being presented with and overseeing the weighing of various goods from foreign lands, including Punt, which in this case included ivory and ebony (wood products).

Another depiction of Puntites, dated to around the same period, appears in the tomb of Rekhmire, who served as vizier during the reigns of Thutmose III and his successor Amenhotep II. In the uppermost part of the scene, Puntites are shown carrying myrrh trees, elephant tusks, gold, animal skins, baboons, and other goods.

Amenhotep II

A tomb of unknown ownership, dated to around the reign of Pharaoh Amenhotep II(~1425-1400BC)97, contains a scene depicting the arrival of Puntites by raft, each vessel carrying four sailors. At the bottom of the scene, an Ancient Egyptian is shown riding a chariot after conducting the trade. This provides clear evidence that trade was not one-sided; Puntites also travelled to Egypt and actively participated in initiating exchange.

The inscription at the bottom reads:

Thutmosis IV (1400–1390 BC)

Further evidence of Puntites travelling to Egypt can be found in the tomb of Amenmose, a Treasurer of Pharaoh Thutmose IV(~1400-1390BC98), his tomb depicts an Egyptian official arriving on a chariot to greet a Puntite envoy99.

Donkeys and Ancient Egyptians carrying goods are also depicted, intended for trade with Punt100.

Amenhotep III

Further evidence of Puntites travelling to Ancient Egypt for trade can be found in the tomb of an official named Sobekhotep (also known as Panehsi), dated to the reign of Amenhotep III (~1390BC-1352BC101). An inscription records that he travelled to the Egyptian seacoast (likely the port of Quseir/Mersa Gawasis), where he met a Puntite envoy who presented him with incense.

The inscription is as follows:

Late Punt (1350BC-1000BC)

During the Eighteenth Dynasty, the Pharaoh Horemheb (~1323–1295BC102), recorded the reception of a Puntite envoy at the Temple of Karnak. The inscription reads as follows:

Like the claim made by the Chief of Punt during the expedition of Pharaoh Hatshepsut, this statement that “our fathers have not trodden it” is clearly hyperbolic and not accurate. As discussed earlier, there is recorded evidence of Puntites travelling by sea and meeting the Pharaoh prior to this period.

Seti I & Ramses II

The temple at Abydos contains two inscriptions referring to incense being brought from Punt, one of which dates to the reign of Pharaoh Seti I (~1294–1279 BC103). The inscription states the following:

Another inscription from the time of Pharaoh Rameses II(1279–1213BC104), mentions Punt:

Ramses III (1217BC-1155BC)

In the Papyrus Harris I, which records the reign and achievements of Ramesses III, there is a passage describing a voyage to Punt. This expedition was large in scale and resulted in a large collection of myrrh, and it even included children from the Puntite chieftain’s family accompanying the return journey to Egypt. This represents one of the final known references to Punt.

The inscription reads:

Conclusion

After the end of the New Kingdom period (~1050 BCE), references to Punt largely disappear from Egyptian records, and it gradually becomes a more mythical, distant land associated with luxury goods such as incense and myrrh.

This decline in references coincides with the Third Intermediate Period, when political power in Egypt became increasingly decentralised. Authority shifted away from the Pharaoh and into the hands of regional rulers, particularly the High Priests of Amun. Unlike earlier periods of fragmentation, this era did not culminate in a strong reunification, instead, it ultimately led to the Persian conquest around 525 BCE105.

It was not until the arrival of Alexander the Great and the establishment of the Ptolemaic dynasty under Ptolemy I Soter that Egypt regained the political strength and maritime organisation necessary to resume proper Red Sea trade. This helps explain the scarcity of evidence for expeditions to Punt during the Third Intermediate Period. Under rulers such as Ptolemy II Philadelphus and Ptolemy III Euergetes, Red Sea commerce was actively revived, contributing to the rise of key trading centres such as Adulis, which I discussed in my article on Adulis.

However, civilisation didn’t totally collapse in the region. Instead, a new polity, the Kingdom of DʿMT (ደዐመተ), emerged in parts of present-day southern Eritrea and Tigray in the early first millennium BC. Rather than ties with Egypt, DʿMT had links with the Sabaean Kingdom to the east, which I explained in further detail in my article on DʿMT.

Hopefully, in the future, more archeological research in Ancient Egypt, but especially around Eritrea, Djoubiit, Somalia and the surrounding regions can shed more light on the Land of Punt and its history.

….

Further Reading

My favourite sources when researching Punt were:

Profiling Punt : using trade relations to locate ‘God’s Land’GOD’S LAND’
Probably the best academic source on Punt currently available, goes into detail into the historical timeline of Punt.

https://www.academia.edu/9111021/Where_is_Punt
Where is Punt? By Charlotte Morgan - Great look into where Punt could be and it’s general history.

https://landofpunt.wordpress.com/tag/ancient-inscriptions/

Although not technically a academic source, it’s heavily citied and is a great introduction into the land of Punt.

Date Created: 1/4/2025

Last Updated: 1/4/2025

Introduction

At the dawn of the 3rd century AD, the earliest Aksumite ruler attested in historical records emerges. He is named as Emperor GDR (possibly vocalised Gadar or Gadarat1 or Gedur, Gadura, Gedara2) in the inscription from Addi Gelemo in Tigray. While in the inscriptions discovered in Yemen, his name is GDRT (possibly vocalised Gadarat3). These are the same person4.

During his reign, he became deeply involved in conflicts across South Arabia. In the north, around Najrān5, and further south near Ṣanʿāʾ6, he confronted the forces of Saba. In the west, he occupied the Himyarite capital of Ẓafār7, while to the east, he supported the Sabaeans in their struggle against Ḥaḍramawt8.

As will become clear later in this article, his engagements were not confined to military confrontation alone. Emperor GDR also pursued an active diplomatic policy, brokering alliances with rival South Arabian kingdoms when it suited his interests and dispatching delegations as part of his political strategy9.

The involvement of the Habeshas predates the beginning of the third century AD. As discussed in my earlier article on Monumentum Adulitanum II, the Robin–Umm Layla 1 inscription, dated to around AD 160, already records engagements with the Habeshas. However, since the inscription is unlikely to involve GDR and is more closely aligned with the period of the Momentum Adulitinum II, it will not be revisited in this article.

Indigenous Inscription

In 1953–54, at Addi Gelemo in eastern Tigray, a bronze sceptre-like object, measuring 55 cm and weighing 455 grams, was discovered10. The piece bears an inscription in unvocalised proto-Geʿez that mentions a king of Aksum named GDR.

The inscription on the sceptre is as follows:

There have been several different interpretations by scholars regarding the deciphering of the inscription on the sceptre. I will first present some of them and then make an educated guess on what it might have meant.

Interpretations

A. J. Drewes & A. Caquot

Manfred Kropp

Alexander Sima

My opinion

To decipher the inscription, we must first break the sentence down word by word and establish the meaning of each term.

• GDR → is likely a reference to the same king GDRT known from South Arabian inscriptions dated to the early 3rd century AD. He may also correspond to the ruler recorded in several Ethiopian king lists under names such as Agdar, ʿAgdar, ʿAgdor, ʿAgdur, or Za-Gadur/Gadur11. Notably, in some regnal/kings list traditions, Agdur appears as a predecessor of Abreha and Asbeha (Ezana and Saizana)12, which aligns well with the proposed historical placement of GDR roughly a century before the reign of Ezana and Saizana.

These king lists are heavily distorted due to the long periods they reach back into the past and the numerous times they have been copied, however, a kernel of historical truth can possibly still be found…

• ngsy → This one is straightforward: it is the Geʿez term ነጋሢ (nǝgśi), meaning “king”13.

• ʾksm → is another straightforward term, simply referring to the city of Aksum14.

• tbʿl → Scholars such as Manfred Kropp, André Caquot, and D. A. Johannes have theorised that this word derives from a root meaning “to possess”, comparable to Hebrew baʿal (“master of”, “to have control over”) and በዓል (bälaʾ) in Geez15, which likewise means “master of”16.

• mzlt → Here, Manfred Kropp suggests that the term refers to the object itself, comparable to Sabaean ZLT, which can denote a votive object17. This is also the conclusion reached by Abraham J. Drewes.

• lʾrg → There are two main theories. The first holds that lʾrg refers to a place-name. During the campaigns of Ezana and Saizana in the early 4th century AD (DAE no. 9)18, conducted against the Agʷēzāt, a locality named እርግ (ʾEreg [?]) is mentioned. Some scholars have hypothesised that this may be identical with lʾrg19. The location of the Agʷēzāt isn’t completely known; however, some scholars theorise that this might be a term for the Agazi20, hence the location of lʾrg might be in southern Eritrea or northern-eastern Tigray.

  The correlation of Agʷēzāt with Agazi is far from certain21, in-fact it’s possible it’s in reference to a location near the Agaw people22.

• wllmq → This term doesn’t appear in other inscriptions; therefore, the meaning isn’t certain. However, some scholars theorise that it is likely a toponym23, since it appears contextually alongside lʾrg, which itself is probably a place name. However, A. Jamme proposed that it refers instead to the god Ilumquh, a suggestion with which I disagree with as it’s never attested to in the form wllmq24.

After analysing each word individually, a plausible conclusion emerges: the inscription on the sceptre likely records a dedication by an Aksumite king named GDR to the important locations of lʾrg (situated somewhere in the Agazi/Agaw lands) and wllmq (unknown location).

Therefore, I propose the following:

Another plausible interpretation might be that of scholar Alexander Sima:

South Arabian Inscriptions

CIH 308/CIH 308 bis

The earliest South Arabian inscription referring to GDR dates to around 200 AD and was discovered at Riyām, in present-day Yemen (CIH 308 and CIH 308 bis). In total, sixteen closely related inscriptions have been identified at the site25. Most are fragmented and difficult to reconstruct; however, two survived in a comparatively well-preserved state.

These inscriptions were composed during the reigns of Sabaean King ᶜAlhān Naḥfān & his sons ŠaʿirumʿAwtar and YarīmʿAyman26. The inscription records, among other events, the establishment of a treaty with GDRT, the king of the Habshites (mlk Ḥbstn).

The inscription reads as follows:

Analysis

From this inscription, we gain valuable insight into the reign of Emperor GDR around 200 AD. His involvement in South Arabian politics implies that he had already consolidated control over his own dominion (the northern highlands of Ethiopia and Eritrea). Furthermore, his command over Adulis is evident, as a substantial naval force would have been necessary to transport troops across the Red Sea. This aligns with the conquests described in the Monumentum Adulitanum, which likely took place only a few decades before 200 AD.

We can also infer that GDRT was a pragmatic ruler who valued diplomacy. Lines 10 and 11 of the inscription indicate that it was GDRT himself who initiated a diplomatic mission to seek an alliance, and that this treaty served both offensive and defensive purposes.

Lines 13 and 14 refer to Silḥīn (possibly vocalised as Salhen), the palace of King ᶜAlhān Naḥfān in Maib, Yemen, and contrast it with the palace of GDR, located in a place called Zrrn (possibly vocalised as Zararan), presumably somewhere within Aksum or its periphery27.

NNAG 13+14

Chronologically, the second South Arabian inscription datable to the reign of GDRT is NNAG 13+1428. Although brief, this inscription is significant in that it reflects a Himyarite perspective. At this time, Himyar was ruled by the dū-Raydān tribal group, and in inscriptions of the 3rd century AD, the polity of Himyar is frequently referred to simply as dū-Raydān29.

Analysis

The analysis of this inscription is brief. During the reign of GDRT, as attested in CIH 308 / CIH 308 bis, an alliance was concluded with Saba under ᶜAlhān Naḥfān, king of Saba, around 200 AD. This inscription demonstrates that the alliance was subsequently put into effect, with a joint military campaign launched against Himyar in the south. The king of Ḥaḍramawt, along with several other tribes, also appears to have participated in this coalition against Himyar.

Lines 3–4 indicate that these combined forces were ultimately unsuccessful and driven back from the Himyarite territory.

A key question is why Emperor GDR allied with the Sabaeans against the Himyarites. The answer is almost certainly economic. The Himyarites controlled the coastal maritime trade on the eastern side of the red sea, which the Aksumites wanted to monopolise. Aligning with the interior Sabaeans against Himyar was therefore a strategic choice.

Ir 12

Another little-known text, Ir 12, is dated to the era of Emperor GDR (during the reign of ŠaʿirumʿAwtar - son of ᶜAlhān Naḥfān30). It records a contingent of Emperor GDR’s army operating on the borders of Ḥashid (north of Ṣanʿāʾ31) and fighting there against the Sabaeans.

The inscription reads as follows:

Analysis

The conflict appears to have flared at least twice: peace was brokered temporarily, then war resumed, with 2,500 Habashite troops and some Bedouin allies raiding the valleys west of Hsdm. This provides evidence of Emperor GDR likely controlling territory much further south than Narjan, possibly having a strong influence in the territory between Ṣanʿāʾ & Narjan.

Ry 533

Ry 533 is a little-known inscription from the era of Emperor GDR that records the Aksumites assisting Saba in a war against Ḥaḍramawt, specifically during the reign of Shaʿrum Awtar (~215–225 AD32).

The inscription reads as follows:

Analysis

It is important to note that detachments or auxiliaries are explicitly mentioned in connection with Emperor GDR’s army. This most likely refers to an indirect support provided to the kingdom of Saba during the reign of Shaʿrum Awtar, during their campaigns against Ḥaḍramawt, rather than a direct Aksumite invasion.

JA 635

JA 635 was inscribed during the reign of Shaʿrum Awtar (~215–225 AD33), king of Saba. In it, he recounts a battle against the Habeshas in Narjan. This is particularly significant, as Narjan was historically an Aksumite stronghold, suggesting that by the early 3rd century AD, the Aksumites had already controlled this strategic gateway to the Tihama trade routes.

The inscription reads as follows:

Analysis

There is little more to add, other than that this evidence demonstrates beyond any reasonable doubt that, during the reign of Emperor GDR, the Aksumites exercised control over Najrān and its surrounding regions. This aligns well with the earlier inscription (Ir 12), which already places them in conflict with the Sabaean Kingdom much further south, near the northern approaches to Ṣanʿāʾ.

Control of Najrān lends strong support to the likelihood that the Aksumites also dominated much of the Tihāma coastline, a conclusion that aligns closely with the conquests described in Monumentum Adulitanum II.

JA 631

Another South Arabian inscription referencing Emperor GDR dates to the reign of the Sabaean king of Sumhuram, Yhwld, who ruled around 230 AD34. It records a delegation sent to the Aksumite Empire during the reign of Emperor GDR, though it appears this mission failed to achieve its intended goal, as the inscription later mentions a war being waged against Aksumite forces.

The inscription reads as follows:

Analysis

This inscription, discovered in the temple of Maḥram Bilqīs at Marib35, Yemen, offers several valuable insights into the reign of Emperor GDR. Firstly, the text identifies him not only as the King of the Ḥabashat (mlk Ḥbst) but also as the King of the Aksumites (mlk ʾksmn)36. This dual title indicates that by at least around 230 AD, the Aksumites had emerged as the dominant power among the Habesha peoples of the northern Ethiopian and Eritrean highlands.

Mahram Bilqis means Sanctuary of Bliqis, Bliqis (another name for the Queen Of Sheba)37.

It also highlights a distinction between the two terms at this time, Habesha appears to have functioned as a broader, more inclusive term encompassing various Semiticised populations of northern Ethiopia and Eritrea, whereas Aksumites referred more specifically to the population centred on Aksum and its immediate periphery.

Secondly, the inscription records that a diplomatic mission was dispatched from the kingdom of Saba to Aksum, likely to negotiate a new alliance similar to that previously arranged during the reign of Alhan Nahfan, as discussed earlier. However, whatever agreement was reached soon collapsed38. The subsequent events described in the text state that GDRT’s son, BYGT (possibly vocalised as Beyga or Baygat39), captured Ẓafār, the capital of the Himyarite Kingdom and established an Aksumite encampment in the centre of the city. In response, the Sabaeans launched a counteroffensive, initiating a fierce engagement that lasted more than three days.

According to the inscription, the first and second days of battle ended with the Aksumites being routed. In the early hours of the third day, the Sabaeans launched a surprise assault, resulting in the deaths of over 400 Aksumite soldiers40. This figure provides a rough indication that the Aksumite army at this time might have numbered several thousand. The fighting continued through the third day in close combat until, by the fourth day, the Aksumites, who were cut off from trade routes and provisions, were forced to withdraw from Ẓafār to a nearby location referred to as Mʿhrtn41.

These conflicts were not confined to land. In fact, it is highly likely that many Aksumite engagements were naval42. Line 7 provides direct evidence of this from the Sabaean perspective: “from the coast and from the land”.

Another interesting point is that GDR likely participated personally in these campaigns in South Arabia43.

Last Updated: 10/11/2025

This article is still in draft phase and won’t be finalised for some time. If you spot any errors or have additional information, feel free to contact me here, or on TikTok or Reddit, along with your sources, so we can discuss

ደግያት/däggiyat is a shortened form of the military title ደጃዝማች/Däǧǧazmač and was one of the highest military ranks in present-day Eritrea & Ethiopia during the medieval era1.

Childhood

ባህታ → Bahta in Tigrinya means joyful or enjoyment, while ሓጎስ → Hagos ሓጎስ also means enjoyment or happiness, making Bahta Ḥagos a truly joyful name indeed.

Bahta Ḥagos was born between 1839 and 1850 in Säğänäyti, a town in the historical province of Akkälä Guzay (present-day Debub/Southern region of Eritrea). He was the son of Shum Ḥagos Andu (ሓጐስ ሹም ዓንዱ) and Wäyzäro Wnahu (ወይዘሮ ውናኡ), the daughter of Shum Mosahzgi (ሓጐስ ሞሳዝጊ)2. It’s said that Bahta’s grandfather, Andu Habal, had become wealthy from bringing salt from the lowlands around Massawa3; neither of them was part of the nobility4, however. Bahta’s lineage traced back to the humble house of ʿEgälto, a branch of the Ṣanʿadäglä descent group.

Video showcasing Säğänäyti in 2024. Source: MicaelGoitom on Twitter

Mohammed Nur Abdu, in Political History of Akkälä Guzay: 1885–1895, records his birth as 18515. Yet an earlier date in the 1840s seems more plausible, photographs from the 1890s depict him as a man in his early fifties. This remains my own speculation, though the Italian author Rosalia Pianavia-Vivaldi likewise suggests a birth in the 1840s.

Bahta’s Siblings6:

• ተኪኤ - Tekelyeh

• ካሕሱ - Kahsu

• ሰንጋል - Sengal

• ግምጃ - Gmjah

• ደስታ - Desta

Although he lived his childhood as a peasant, working his father’s land and herding goats and cattle like many highlanders of the time, local tradition recalls that he was renowned from an early age for his courage and strength. One story tells of a traveller passing through Säğänäyti who challenged anyone to a fight; while the other children stayed silent, he alone stepped forward7.

During his late teenage years, Bahta Ḥagos encountered a major challenge when his father, Ḥagos Andu, aligned himself with the Hamasien ruling family of Märäb Mǝllaš, the House of Šäʿazzäga. This alliance was driven by his friendship with the Kantiba of Karnishem (a town in the historical province of Hamasien - present-day Central region of Eritrea), who had been appointed by Ras Ḥaylu of Šäʿazzäga8, and by his hope that supporting Ras Ḥaylu would secure him a similar title in Säğänäyti. However, the šmagles (elders) of Säğänäyti chose instead to back the traditional House of Ḥazzäga, which upheld their democratic system and the independent authority of Akkälä Guzay.

In 1861, Däǧǧazmač Ḥaylu of Šäʿazzäga was the ruler of Märäb Mǝllaš. However, Ras Wäldä Mikaʾel Sälomon of Ḥazzäga, who governed a portion of Hamasien under Ras Ḥaylu’s authority, later rebelled during the late 1860s9. Following his revolt, Wäldä Mikaʾel fled south of the Mereb River, where he lent support to Däǧǧazmač Kaḥśay (the future Emperor Yohannes IV) in his campaign to overthrow Emperor Tewodros II10.

As a result, Ḥagos Andu was condemned to death in a Säğänäyti Bayto session11. The loss of his father at this critical stage undoubtedly left a deep mark on Bahta Ḥagos and may in part explain his later ambitions for power.

The Killing Of Fitawrari Émbayye

Fitawrari was a military rank in medieval Eritrea and Ethiopia. A Fitawrari was responsible with overseeing the army’s supply, maintaining order and discipline while on the march, and ensuring the overall safety of the troops. The Fitawrari commanded the forward position of the army (the advanced guard)12.

The first major event for which Bahta Ḥagos is remembered was his involvement in a dispute with Fitawrari Émbayye, the son of the then-governor of Akkälä Guzay, Arʾaya Śəllase Dəmsu13. This conflict arose as a direct consequence of the excessively harsh taxes imposed by Arʾaya Śəllase Dəmsu after his release from imprisonment in Mäqdäla, where he had been held until Emperor Yohannes IV’s successful rebellion against Tewodros II14.

In October 1875, at Säğänäyti, soldiers of Ras Arʾaya Śəllase Dəmsu attempted to levy taxes on the townspeople. This provoked a clash in which several individuals on both sides were wounded or killed, among them ጓንጉል/Gwangul, the uncle of Bahta Ḥagos. Enraged by his uncle’s death, Bahta intervened and, in the heat of the conflict, killed Fitawrari Émbayye15. Over 20 of Émbayye’s soldiers were killed, while eight men of Säğänäyti were dead16.

This occurred during the period of the Egyptian invasion of Märäb Mǝllaš, marked by the Battle of Gundet (14–16 November 1875)17. It may have been these turbulent conditions that drove Bahta, his brothers, and other notable figures of Säğänäyti to take such a drastic step of rebelling and seizing the moment to kill Fitawrari Émbayye…

Although the death of Fitawrari Émbayye undoubtedly grieved his father, Ras Arʾaya Śəllase Dəmsu, he nevertheless pronounced a verdict of mercy18 towards Bahta Ḥagos and the others involved in his son’s killing. This was likely because of the rather harsh measures Émbayye had inflicted on the Säğänäyti people and the number of lives he had taken or harmed.

Life as a Šafta (Rebel)

Šafta19 are outlaws who typically lived on the periphery and harassed local rulers by raiding or attacking merchants, what differintiates them from banditary is the more political aims of the Šafta, Šafta’s were alot of times political exiles.

In September 187620, Emperor Yohannes IV directed Ras Alula to confront Ras Wäldä Mikaʾel Sälomon of Ḥazzäga and depose him from authority in Märäb Mǝllaš. To carry out this mission, Ras Alula passed through Akkälä Guzay, where he plundered the property of Bahta Ḥagos and his family in Säğänäyti21, famously taking Bahta Ḥagos’s favourite cow. Säğänäyti itself suffered the wrath of Alula22.

At this point of time, Märäb Mǝllaš was in state of chaos, Ras Ḥaylu of Šäʿazzäga was killed by Ras Wäldä Mikaʾel Sälomon of Ḥazzäga, when the later confronted Haylu’s army at the battle of Wäkkädäbba on 17th July 187623, pre-empting the actions of Emperor Yohannes IV stated above…

Therefore, to avoid further persecution of themselves and the people of Säğänäyti, Bahta Ḥagos and his brothers fled to the Asaorta Saho lands, north-east of Säğänäyti, and eventually settled in the town of Agämäda24. There, they became Šafta from 1876 to 1880. They began amassing loot, including firearms, through the forceful coercion of passing caravans.

However, the Šafta activities seem to have taken a toll on Bahta Ḥagos. Seeking another way to regain control of his homeland Säğänäyti, he clashed with his brother Kahsu, who wanted to keep raiding caravans to stockpile arms for a later, forceful conquest. The dispute escalated, and Bahta told his brother he would either kill him or surrender him to Alula to have him killed. Kahsu replied, “Go and surrender to Alula”. Bahta prayed through the night and, at dawn, killed his brother. The act haunted him for the rest of his life25.

Some sources mention that Kahsu wanted to surrender to Alula, which angered Bahta greatly26.

Bahta Ḥagos & The Habab

ከንቲባ/Käntiba was a term used to denote a governor of a large town/city, usually highly autonomous and self designating but not always27.

By 1880, Ras Alula’s lieutenant, Bälatta Gabru, was tasked with expelling Bahta Ḥagos from Akkälä Guzay, a mission he successfully carried out28. Bahta Ḥagos then moved north into the land of the Habab29 (in the Sahel region of Eritrea, northeast of Keren, in today’s Northern Red Sea Region, specifically the Nakfa Sub-zone30). The Habab Käntiba Hamid Hassen granted him refuge from Ras Alula’s pursuit, but even there, he was not safe. Ras Alula repeatedly attacked the Habab, regarding the region as the rightful territory of the Ethiopian Empire. In October 1880, for instance, Alula launched a raid with an army of more than 12,000 men31. Alula eventually withdrew to Šäʿazzäga in Hamasien after Rashid Pasha and Bahta Ḥagos assembled some 7,000 riflemen and prepared to strike from the rear, with the support of their garrison at the fort of Sanhit (at Keren)32.

A lesser-known reason for Bahta Ḥagos’s alliance with the Ḥabab likely stems from genealogical ties. As previously mentioned, Bahta Ḥagos was from Säğänäyti, and the Ḥabab (together with the Ad Täklə and Ad Təmaryam) form part of the Bet Asgäde. All three groups trace their lineage to a 16th-century ancestor named Asgäde, who is said to have originated from Säğänäyti, a tradition that remains well remembered in the town to this day33.

The relationship between Käntiba Hamid Hassen & Bahta Ḥagos was said to be extremely close, with Käntiba Hamid Hassen giving him the title of Käntiba34.

On 16 January 188235, Ras Alula raided the region around Debre Sina and imposed tribute, which was then delivered to the monastery at Debre Sina (just east of Keren). As his troops withdrew, Bahta Ḥagos and Sängäl, in defiance of the tribute, attacked and looted the monastery, leading to the deaths of several monks36. At the same time, Rashid Pasha ( Egyptian Commander at Massawa) retaliated by burning villages that had paid Alula’s tribute. This cycle of reprisal continued, and in April 1882, Ras Alula returned to Debre Sina to pursue the Habab and suppress their resistance in the area37.

Ras Alula is said to have requested Rashid Pasha to capture Bahta Ḥagos, but the request was refused. Instead, Alula repeatedly conspired to seize him through espionage38, however this wasn’t always the case as Muzinger (Governor of Massawa & its environs for a period of time) had looked the other way while Ras Alula conducted raids into Habab territory trying to capture Bahta39.

Bahta Ḥagos & The Italians

Historical Background

It’s crucial to understand the circumstances that led Bahta Ḥagos to initially side with the Italians before passing posthumous judgement. Historical figures of that era faced complex circumstances, and a nuanced perspective is necessary, for no one is without flaw…

With the collapse of Egyptian control and influence in Massawa, Senhit, Habab, and the surrounding regions between 1882 and 1885, a power vacuum gradually emerged, soon to be filled by the Italians. Advancing northwards from their colonial foothold at Assab in southern Eritrea, and aided by the Afar Sultan Mohamed Hanfare, the Italians gained momentum. On 25 January 1885, Egypt signed a secret agreement with Italy to withdraw from the historic port of Beilul in southern Eritrea40. The following month, in February 1885, Italian forces under Admiral Caimi occupied the port of Massawa41.

During this period, Akkälä Guzay remained in turmoil, positioned beside the Egyptian and later Italian-controlled coastline. The region became a battleground between these powers and the forces of Emperor Yohannes, particularly those led by Ras Alula. In 188542, for instance, Ras Alula launched a campaign against Catholic missionaries, which undoubtedly affected sections of Akkälä Guzay’s Catholic population.

From Alula’s perspective, this move sought to consolidate authority over Akkälä Guzay, a strategically important frontier province that bordered both Egyptian and later Italian influence. Jesuit correspondence shows that Alula could act cordially toward Catholic residents, provided they agreed to pay tribute and refrained from separatist activities43.

Interactions with Jesuits predate the 19th century. During the early 17th century AD, Bahr Negasi Yohannes Hakay44 permitted Jesuits to enter Bur to seek refuge after they were expelled by Emperor Fasilides. Still, the majority of Säğänäyti’s population remained Orthodox until the 19th century AD.

Ras Alula would be stripped from his governorship of Akkälä Guzay in 187645.

Around May 1888,46 Däbbäb Ar’aya (2nd son of Ar’aya Ṣəllase Dəmṣu) was appointed governor of Akkälä Guzay by Emperor Yohannes after spending years as a šafta in the region and acting as an ally of the Italians (though he eventually defected47). On 9 March 1889, during the Battle of Metemma, Emperor Yohannes IV and several important nobles, including Ras Ar’aya Ṣəllase Dəmṣu & Däǧǧazmač Gabr Farus (Governor of Akkälä Guzay)48, were killed in the ensuing chaos.

With Italian arms and backing49, Däbbäb Ar’aya rebelled against Ras Mengesha (son of Emperor Yohannes IV and heir). By 3 July 1889, he had encamped at Adwa, and between 17 and 19 July, he reached Mekelle, where soldiers of Ras Alula arrested him50. This set of events left Akkälä Guzay in a state of turbulence, and it was in these circumstances, Bahta Ḥagos entered negotiations with the Italians at Massawa in 1889.

The Italians were actively seeking to recruit Eritrean nobles to support their drive to extend colonial control into the highlands. With the region already in deep turmoil militarily, economically and socially because of the famines of the late 1880s and 1890s, it is little wonder that many nobles of Märäb Mǝllaš turned to Italy.

Bahta Ḥagos and his entourage were warmly received by the Italians. Bahta himself was granted the rank of ደጃዝማች (Däǧǧazmač), his brother Sängäl was appointed ኣዝማች (Azmač – commander), and his son Gäbrämädhén received the title ግራዝማች (Grazmač – commander of the left flank)51.

On 2 May 188952, Emperor Menelik II concluded the Treaty of Wuchale with Italy, formally setting the boundary between the Ethiopian Empire and the new colony of Eritrea. The agreement placed Asmara & Halai as the colony’s edge53. Between 2 and 4 August 188954, Italian forces occupied Asmara with the support of Bahta Ḥagos, who protected their flank with three hundred men.

Still unsatisfied with their territorial gains, the Italians pushed further into Seraye and Akkälä Guzay, advancing as far as the Mareb River, regarded as the “natural frontier” of Eritrea. By February 1890, the whole of Akkälä Guzay, including the Shimezana district (bordering Agame), had fallen under Italian control, and Bahta Ḥagos was promoted to Däǧǧazmač of the province55.

Däǧǧazmač Bahta Ḥagos eventually clashed with Däǧǧazmač Sebhat Arigawi56, the provincial ruler of Agame. Their feud was soon halted when Arigawi sought a truce, as the great famine of the 1890s was at its peak. Unlike the Eritrean colony, which could draw some relief from the Italian mainland, the Tigrayans and the rest of Ethiopia had no such support and were severely affected.

By the early 1890s, Däǧǧazmač Bahta Ḥagos had gathered well over 500 troops, with some accounts noting that he possessed more than 2,000 rifles, and had become the most powerful indigenous leader, particularly after the Italians killed Kefla Yasus, the son-in-law of Ras Wäldä Mikaʾel Sälomon57. He also accumulated considerable wealth, with one-tenth of the tax from Akkälä Guzay going into his pockets58, with one source stating he had around 600,000 lira59 (Italian currency pre 2002).

The strength of Bahta’s army was evident when they served as part of the Askari contingent alongside the Italians60 at the Battle of Agordat on 21 December 1893 against the Mahdists, a battle they won despite being heavily outnumbered.

Rosalia Pianavia, the wife and biographer of Colonel Domenico dei Marchesi (the senior Italian officer in Asmara61), recalls an encounter with Bahta Ḥagos. One day, Bahta arrived at her home accompanied by his brother Däǧǧazmač Sängäl, his son Gäbrämädhén, several chiefs and sub-chiefs, and a hundred armed men. He rode a white mule draped in a red, silver-embroidered cloth62.

Yet Bahta did not command universal support. Kantiba Metalka of Maratta-Sabana, south of Säğänäyti, famously declared, “I do not know any Bahta here; I do not recognise anyone above me”. This confrontation was eventually defused when the Kantiba’s daughter married Bahta’s eldest son. Bahta’s eldest daughter, Tamnit, later married the son of Kantiba Asmrmon of Maraba, west of Säğänäyti, while his second daughter, Lamlam, married Menelik Sebhatu of Karnishem in Hamasien63. All of these sons-in-law were required to convert to Catholicism.

When Däǧǧazmač Bahta Ḥagos first entered Säğänäyti, resistance was minimal. In fact, most of the Akkälä Guzay nobility rallied to his side64. However by መስቀል (Mäsqäl - Feast of the cross celeberation) 1889, resentment by some Säğänäyti nobility was already brewing, leading to some arrests by Bahta Ḥagos65 & even a battle against some Asaorta Saho who refused to recognize his authority66.

Bahta Ḥagos was far from perfect. During his rule, there was evident favouritism towards the people of Säğänäyti, while several other villages suffered under high taxes imposed by his subordinates. For instance, his son and close associates were said to have been particularly harsh in collecting taxes in the village of ዳርጋ → Daber67.

The Rebellion

In June 1893, the Italians under General Oreste Baratieri disbanded Däǧǧazmač Sebhatu of Karnishem and his army. This incident, combined with the gradual erosion of the indigenous nobility, the loss of native land rights in the colony, and the widespread sexual abuse of women (such as the use of local women in Italian military brothels68), strongly influenced Däǧǧazmač Bahta Ḥagos to rise in rebellion against the Italians69.

Even prior to this, tensions were rising between the Italians and Bahta Ḥagos. They repeatedly summoned him and his retinue, warning that any act of rebellion would be crushed. Yet they hesitated to arrest Bahta, wary of the uproar such an action would provoke70.

Some Italian spies reported that Bahta Ḥagos was collaborating with Ras Mengesha, the son of Emperor Yohannes and then ruler of Tigray, in planning a joint attack against the Italians. However, there is no substantial evidence to confirm that such coordination ever took place71.

On December 14, 189472, a dramatic event unfolded at the residence of Giovanni Sanguinetti, the governor of Säğänäyti. Bahta Ḥagos, accompanied by his brother Sängäl and his son Gabra Madhen, arrived for their usual coffee. However, the visit was a ruse. After finishing their coffee and liquor, Bahta Ḥagos and his associates suddenly grabbed the governor. They threw Sanguinetti to the ground and bound him, with Bahta Ḥagos pinning the governor’s head with his knee73.

Then an interesting dialogue occurred between Batha Hagos & Sanguinetti:

Gabra Mika’el Germu, an intellectual and interpreter serving the Italians, provided additional dialogue attributed to Bahta Ḥagos74…

Bahta Ḥagos and his associates then severed the telegraph lines to Asmara, effectively cutting off all communication between Säğänäyti and the Italian headquarters. Gabra Egzi’abeher, the interpreter for the Italians, was offered assistance to care for his blind mother in Asmara and promised a prestigious title if he would cease aiding the Italians and instead join Bahta’s side75. However, Gabra refused. Enraged by this, Bahta Ḥagos exclaimed the following:

If the above quote is indeed authentic, it suggests that Bahta believed he would receive support from Ras Mengesha and Emperor Menelik, as well as from the local Asaorta Saho and possibly the Muslim communities. However, it is also likely that Bahta made this claim merely to persuade Gabra Egzi’abeher to join his cause and offer assistance.

On the night of December 14th 1894, he issued the following proclamation encouraging all those in Akkälä Guzay and in colonial Eritrea in general to rise against the Italians:

However, some remained hesitant and not fully committed, including his brother Sängäl and his son Gäbrämädhén. Which prompted Bahta Ḥagos to lament the following memorable quote76:

There is also a slightly longer version, referencing the arrival of Italians at Assab & Massawa…

The Battle of Halai

On 15 December 1894, General Oreste Baratieri ordered Major Pietro Toselli to march with his askari forces into Akkälä Guzay to suppress Bahta Ḥagos’s rebellion. Toselli reached Säğänäyti on 16 December, where he entered into negotiations with Bahta Ḥagos, who had gathered an army of roughly 2,000 men.

The discussions soon deteriorated, and by 18 December, Toselli commanded his troops to advance into Säğänäyti late in the afternoon. By then, however, Bahta Ḥagos and his forces had already withdrawn north-eastwards to the village of Halay, aiming to secure the stockpile of armaments stored there and take control of the strategically positioned fort.

At the Fort of Halai, around 250 askaris were stationed under the authority of Captain Castellazzi77 & Lieutenants Soliani and Basile78, in a well-defendable position. Bahta Ḥagos sought to negotiate with the Italian commander through two Catholic priests, Abba Kefla Maryam and Abba Takla Haymanot79, but little progress was made. The Italian officer was merely stalling for time, awaiting the arrival of Toselli’s reinforcements from the flank.

Possibly realising this deception on the afternoon of 18 December, at around 1:30 PM80, Bahta Ḥagos gave the order to assault the fort. Despite the defenders’ advantage of firearms and fortifications, Bahta’s men gradually gained ground and appeared on the verge of capturing the fort, as they had surrounded it.

However, before they could prevail, at around 4:45 PM81, Captain Toselli and over 1,200 askaris arrived from the flank and deployed a barrage of artillery.

Bahta had ordered prior to this battle to have all the prisoners, including Sanguineti & Gäbrä Egzi’abeher to be lead away from Säğänäyti, Tesfu Mariam, Bahta’s brother in law would be in charge of this operation82. Eventually however they would agree to release them, if they were promised clemancy from the Italian authorities83.

By 6 PM84, Bahta’s forces were surrounded and were routed, leading to their defeat and the death of Bahta Ḥagos himself, at a place called ዑና ቀራንኤኡ/Una Qarnyehu85, west of Halai near a church86. With 11 askari killed and 22 wounded on the Italian side, and many more on Bahta’s side. Bahta’s brother Sängäl and a smaller contingent escaped southwards to Tigray, where they were given refuge by Ras Mengesha87.

Date Created: 8/20/2025

Last Updated: 8/26/2025

Subscribe now

The absence of definitive evidence confirming whether the emperor referenced in Monumentum Adulitanum II hailed from Adulis or Aksum has led scholars to propose varying hypotheses. As such, there is no single “correct” answer. Beyond modern-day displays of point-scoring (which ultimately hold little significance), the fact remains that this leader emerged from the highland region of present-day Eritrea and Ethiopia. He placed considerable importance on the city of Adulis and played an important role in shaping the history of the Aksumite Empire and, by extension, that of both Eritrea & Ethiopia - Authors Disclaimer.

I encourage you to read the article and explore additional sources to form your own conclusions. Hopefully, future discoveries of artefacts and primary sources will help shed further light on these questions.

Introduction

Around the turn of the 2nd century AD, the highlands of what are now Eritrea and Ethiopia were likely fragmented, with various polities controlling different regions1. Long-established cities such as Yeha and Adulis had already exerted significant influence over the plateau for centuries. Meanwhile, newer cities like Qohaito, Matara & Aksum were beginning to rise in prominence. Aksum, in particular, experienced increased growth during the late 1st century AD. The earliest known reference to the city appears in the Periplus of the Erythraean Sea(40 to 70AD2), which mentions a city of the ‘Auxumites’, which was a settlement located five days inland from Coloe (likely modern-day Qohayto3).

Aksum’s growth during this period was likely driven by several factors, one of which was its advantageous access to trade routes extending deep into the southern and eastern hinterlands of Ethiopia & Sudan. This contrasted with sites like Adulis, which were more distant from these regions. However, there is no conclusive evidence of an established Aksumite kingdom/empire during the 1st century AD. Rather, Aksum was likely a powerful “city-state”, which steadily expanded its wealth and influence throughout the 1st and early 2nd centuries AD. This expansion possibly culminated in conflicts with other pre-existing polities, from which Aksum emerged victorious, becoming the dominant power in the highlands of Eritrea and northern Ethiopia.

As noted in my previous article, there was extensive empire once ruled by Emperor Zoskales, likely reigning at Adulis in the mid-first century AD. However this empire had probably disintegrated by the early second century. Which is why the conqueror commemorated in the Monumentum Adulitanum II had reclaim much of those territories, in addition to newer ones, as we shall examine later in this article.

In terms of archaeological evidence, there is a notable absence of findings directly within Aksum that can be securely dated to the 1st century AD. Even in the surrounding areas, evidence from this period remains limited. However, one significant site just north of Aksum, Beta Giyorgis (roughly translates to House of Saint George), does show signs of occupation during the 1st century AD.

Some scholars have identified Bete Giyorgis as a “proto-Aksumite” site. Specifically at the archeological sites known as Ona Enda Aboi Zewgè (roughly translates to Ruins Of Father Zewgè), several graves containing pottery and stelae have been discovered4. A large building reminiscent of a palace was also unearthed at the nearby site of Ona Nagast (roughly translates to “Ruins Of Kings”), alongside numerous pre-Aksumite and proto-Aksumite elements, including multiple glass fragments dated to around the 3rd century AD5.

Could the palace at Beta Giyorgis have been built for the conqueror recorded in the Monumentum Adulitanum II, or perhaps for his descendants in the third century AD?

Civil War In The Highlands?

Around 143–155 AD (likely 144 AD6, the exact date of The Roman Oration remains debated among scholars7), a possible glimpse into events unfolding in the highlands of northern Eritrea and Ethiopia emerges through the following passage from The Roman Oration by Aelius Aristides.

Some scholars hypothesise that the “wretchedness” described along the Red Sea may refer to civil strife in the region during this period8. Adding to this hypothesis, scholar & historian G. W. Bowersock also notes that a Roman military detachment was sent to the Farasan Islands (off the Arabian coast, directly opposite Eritrea) in response to a “crisis”9, which could be linked to unrest or a civil war in northern Ethiopia and Eritrea affecting the safety of trade in the Erythraean Sea.

A Royal Palace At Aksum?

In my view, one or more wars likely took place during this period between the various highland polities, particularly between Aksum and Adulis, from around 150 AD.10 There are several reasons to support this view. Firstly, Roman sources mention disruptions in the Red Sea and a subsequent Roman intervention. Secondly, the period was followed closely by the rise of Aksum, which undoubtedly led to conflicts with adjacent territories. Lastly, the first section of the Adulis Throne inscription (which we will soon analyse) refers to the conquest and subjugation of several nations in and around the highlands.

Ptolemy’s Geography (written around ~150AD11) mentions both Aksum and Adulis; however, he uniquely notes that Aksum was the site of a “royal palace”. This detail lends weight to the view that Aksum had risen in power by this period, sufficiently so for Ptolemy to single it out as the seat of a king or emperor. In contrast, no such reference to a “royal palace” is made for either Koloe (Qohayto) or Adulis.

This does not imply that royal palaces did not exist in Qohayto or Adulis. As discussed in my other articles, archaeologists have uncovered palace-like structures in both locations. Rather, Ptolemy’s specific mention of Aksum’s royal palace may reflect its greater grandeur, surpassing that of the others, and thus signifying its primacy over the other cities.

However, even if the Aksumites rose to prominence during this period, it does not imply that other identities (such as the Adulites) were entirely erased. This is evident in Ptolemy’s Geography, which refers to both the Adulitae and the Axoumitai as distinct groups. At the very least, this suggests that separate identities between these two populations were still recognised at the time.

In my opinion around 150 AD, the initial conquests described in the Monumentum Adulitanum II were likely underway, marking the unification of the highlands and the probable consolidation of both the Aksumite and Adulite kingdoms, into a single entity: The Aksumite Empire.

War Against Meroe?

Some scholars hypothesise that during the reign of King Sherkarrer (or Shorkaror) of Meroë in the early 2nd century AD, the Aksumites launched an unsuccessful expedition against Meroë12. Supporting evidence comes from a rock carving at Jebel Qeili near modern-day Kassala (Central, Eastern Sudan, close to the border of Eritrea). The carving depicts King Sherkarrer alongside a god-like figure (possibly Apollo, the sun god), with slain enemies beneath them, which some scholars believe might represent Aksumite warriors13.

As we will explore in the next section, the unnamed conqueror referenced in the Monumentum Adulitanum II inscription notably omits any claim of conquering Meroe. This omission was almost certainly not due to ignorance, since several other neighbouring polities are explicitly mentioned, but rather reflects a likely inability to subdue Meroe at the time.

It was not until the late 3rd century AD, during the reign of Emperor Ousanas, that the Aksumite Empire finally succeeded in defeating Meroe.

The Unnamed Conqueror (160–200 AD)

The Throne at Adulis

The reign of the unnamed conqueror is recorded in an indigenous inscription on a marble Throne at Adulis. This throne bears a Greek inscription on both its sides, detailing a ruler who would conquer an Empire spanning a vast region along the coast of the Erythraean Sea. In the 6th century AD, Cosmas Indicopleustes, a Byzantine traveller, mistakenly believed the inscription to be a continuation of a nearby stele associated with Ptolemy III. However, modern scholarship has since disproven this, as discussed in greater detail in my article on Adulis.

The inscription written on the Throne at Adulis is called: Monumentum Adulitanum II (RIÉ 277)

To interpret these campaigns accurately, the first step is to determine the most probable date of the Adulis Throne, thereby establishing when these conquests took place.

Dating the Adulis Throne

The exact date of the Greek inscription on the Adulis Throne remains a subject of scholarly debate. Historian Glen Bowersock proposes that the events described in the inscription occurred in the late 2nd century AD or early 3rd century AD14. I have also concluded that this is the most likely date when the events on the throne occurred. My reasoning is as follows.

The Kinaidocolpitai

The Kinaidocolpitai are a notable group mentioned exclusively between 150 AD and 300 AD15. The Kinaidocolpitai are mentioned in the Adulis Throne & various historical documents, including Ptolemy's "Geography", but they seem to disappear from records by the mid-third century16. This disappearance supports the hypothesis that any inscription mentioning them must have been created during their known period of existence. Thus, an outer and inner boundary in our time frame for the Monumentum Adulitanum II is set: 150AD-300AD.

South Arabian Inscriptions

Secondly, the earliest known South Arabian inscription mentioning the “Habeshas” was discovered at the ʾUmm Laylā fortress in the Yasnam area of Qatābir District, the northernmost governorate of Yemen. It records an invasion by the Habeshas (b-nḥbšn) into Sabaʾ, specifically near Umm Laylā. Scholar Christian Julien Robin dates the inscription to between 160 and 190 AD17, based on references to the Banū Sukhaym tribe, which was ruling Sabaʾ at that time.

South Arabian inscriptions from this period commonly use the term “Habeshas” to refer to foreigners from the highlands of Eritrea and Ethiopia. However, the specific polities that are encompassed by this term remain uncertain, at least until the rise of the Aksumite Empire in the 3rd century AD, when clearer correlations can be drawn.

As we shall see later, the conqueror mentioned in the Monumentum Adulitanum II records Saba as the northernmost limit of his campaigns. Could this suggest he was defeated there, hence the Sabaic inscription known as Robin-Umm Layla 1 referring to the Adulis Throne’s conqueror attacking Saba during the period 160–190 AD?

Inscription Date: 160-200AD

Therefore, based on the analysis of the Sabaic inscription dated to 160–190 AD, the reference to the Kinaidokolpitai, who are attested between 150 and 300 AD and as previously discussed, the Roman sources reporting major unrest and warfare around 150 AD, including the deployment of a Roman detachment to the Farasan Islands in the Red Sea, we arrive at the same conclusion held by most scholars: the inscription likely dates to the late 2nd century AD, specifically from 160-200AD. This view is supported by both scholars, David W. Phillipson18 and Glen Bowersock19.

Who was the Emperor?

The second key question regarding the Monumentum Adulitanum II inscription is the identity of the emperor whose conquests it records. As the inscription does not mention a name, we must rely on educated speculation. Scholars remain divided over who this emperor was and which kingdom he ruled. Based on the previously estimated date of 160–200 AD, no known emperor can be definitively linked to this period. The closest candidate is Emperor GDR (often vocalised as Gadara), who appears in Sabaic inscriptions dating to 200–220 AD.

However, identifying the emperor as Gadara can be problematic, as his earliest known references appear as much as 40 years after the earliest possible date of the first Sabaic inscription mentions Habeshas. It is therefore plausible that Gadara was the successor of the conqueror described in the Monumentum Adulitanum II.

Nevertheless, Scholars such as Glen Bowersock have also suggested that the ruler could have been Gadara20. I’m also of the opinion that it was either Gadara or a direct predecessor.

Aksum or Adulis?

As for the empire to which the emperor belonged, most scholars have traditionally attributed the inscription to the Aksumite Empire (see Munro Hay21, Glen Bowersock22). However, more recent scholarship has proposed that it refers instead to Adulis itself (see David W Phillipson23). These scholars highlight that the inscription was physically located in Adulis and that the emperor explicitly states he "gathered" his troops there, implying that Adulis functioned as the imperial centre.

Researchers such as Fauvelle-Aymar & François-Xavier24 have also pointed to the inscription’s Hellenistic features, including its use of Greek, and the depiction of figures like Heracles (a Greek demigod and symbol of strength) and Hermes (the Greek god of trade and communication, reflecting Adulis’ role as a trading centre). Furthermore, the inscription refers to the conquest of the "Gazē" tribe, which was later equated with the Aksumites in an 11th-century footnote added to a manuscript copy of Cosmas Indicopleustes’ Christian Topography25.

In contrast, proponents of the Aksumite interpretation26 highlight that Aksum was already emerging as a dominant power by the late 2nd century AD and had likely surpassed Adulis in political and military significance. They argue that the conquests described in this were the first step and directly related to the rise of the Aksumite Empire in the 3rd century AD.

The 11th-century copies of Cosmas Indicopleustes’ Christian Topography include a footnote stating that the tribes near Adulis belong to the tribe τιγρητῶν27(tigrētôn), the plural of Tigrētēs. Although this note was written some five centuries after Cosmas, it broadly corresponds to the location of some Tigrē people today, such as the Mäsǝḥalit sub-clan of Tigrē who still live around Zula to this day28.

In addition, certain Saho groups, such as the Iddefer & Asaorta tribes also live around Zula29 30.

My theory

In my view, the ruler in question was part of the Aksumite Empire, though he likely governed from both Aksum and Adulis. I came to this conclusion because of several factors, for example, an inscription dated to 200–220 AD (Ja 631), which identifies Emperor Gadara as an Aksumite. Thus, if the emperor who commissioned the Adulis Throne was Gadara or even his direct predecessor, it leaves little room for a significant shift in political identity in such a short period (<20-40 years), in addition to this the reference of the Aksumite Royal palace by Ptolemy but not any others also leads to thinking Aksum was the strongest power at this time.

However, the reference to gathering at Adulis in the throne inscription, along with its physical placement there & the Hellenistic influence, suggests that Adulis may have served as the emperor’s primary seat of power or at least held a status equal to that of Aksum during the time of the inscription.

This is similar to how Babylon became the long-term capital and residence of Alexander the Great after his eastern conquests and where he eventually died31. A ruler may originate from a foreign land (such as Aksum) yet place particular emphasis on a neighbouring city (like Adulis) because of its historical & strategic significance.

The absence of definitive evidence confirming whether the emperor referenced in Monumentum Adulitanum II hailed from Adulis or Aksum has led scholars to propose varying hypotheses. As such, there is no single “correct” answer. Beyond modern-day displays of point-scoring (which ultimately hold little significance), the fact remains that this leader emerged from the highland region of present-day Eritrea and Ethiopia. He placed considerable importance on the city of Adulis and played an important role in shaping the history of the Aksumite Empire and, by extension, that of both Eritrea & Ethiopia.

Nevertheless, a few key facts about the Monumentum Adulitanum II are clear. The emperor mentioned in the inscription hailed from modern-day northern Ethiopia & southern Eritrea, ruling in the late 2nd to early 3rd century AD and went on to establish the largest empire in the history of the region, extending across both shores of the Erythraean Sea.

The Conquests

Finally, we arrive at the analysis of the conquests detailed in the Monumentum Adulitanum II inscription. Given the length of the text, I will break it down and examine each section individually, including relevant quotes throughout. It is important to note that not all the locations or nations/peoples mentioned in the inscription are identified with certainty. While some, such as Agame, are well known and still exist today, others, like Zaa and Gabala, remain obscure. Where possible, we will rely on historical and contextual evidence to make informed interpretations.

Unification of highland City-States

As discussed in the introductory section of this article, the highlands of Eritrea and Ethiopia during the 1st and 2nd centuries AD were politically fragmented. Since the fall of DʿMT around the 4th century BC, no unified kingdom/empire is known to have existed, besides the Empire of Zoskales (which was largely a coastal empire). Instead, regional polities such as Adulis, Matara, Yeha, and Aksum functioned as independent centres of power.

All of this, however, was about to change, as we will see in the following excerpt from the inscription on the Adulis Throne:

Firstly, it is possible that Cosmas did not copy the introductory section preceding the inscription, which may have included the name of the emperor and his empire. Scholars such as Glen Bowersock support this view32. Regardless, the first surviving line states that the emperor "compelled the nations bordering on my kingdom to live in peace", likely referring to large villages or towns near his nucleus of power (Adulis/Aksum) being subjugated. In either case, securing the periphery would have been essential before conquering territories further away.

Glen Bowersock notes that ἀνδρειώσας (andreiósas) can imply either "manhood" or "manly strength”33. If the former is intended, it may suggest that the conqueror was still quite young at the time of his campaigns.

The first group mentioned are the Gaze (Γάζη), first attested during the DʿMT period around the 6th century BC. As previously noted, this term may refer to the Aksumites themselves (according to an 11th-century marginal note found in a manuscript of Cosmas Indicopleustes34). However, since this note does not appear in earlier 9th-century copies, therefore its reliability is questionable. Even if accurate, it is possible that by Cosmas’s time in the 6th century AD, the Agaze had already been assimilated into the broader Aksumite identity. Another theory proposed by scholars suggests that the Gaze were a people who once inhabited the Akele Guzay region35 & correlate them with the Agwezat, who are later mentioned in an Aksumite inscription as rebelling several times36.

In my view, the Agaze were likely a group originating from the Akele Guzay region of Eritrea and northern Agame area of Tigray, who were later assimilated into the Aksumite identity, possibly through intermarriage with royal lineages in Aksum, thus the late reference in the 11th century copy of Cosmas’s manuscript37.

Interestingly, the late 18th-century explorer Henry Salt noted that the Gaze were located specifically near Adulis, in a region through which significant trade passed into the city.38 This suggests that conquering the Gaze territory was crucial for controlling the flow of goods in and out of Adulis.

Next, the inscription mentions the Agame, a far more straightforward identification. This undoubtedly refers to the Agame region in north-eastern Tigray, which still exists today. As Bowersock notes, it was likely situated near the present-day city of Adigrat in Tigray.39

The final territory mentioned in this initial section is Sigye (Σιγύην, i.e. the people of Sigýē). An 11th-century copy of Cosmas Indicopleustes’ Christian Topography includes a footnote identifying the Sigye as the Suskinitai40, though this offers little clarity regarding their precise identity. However, some possibilities exist. The late 18th-century traveller Henry Salt noted that “Sigue” means “Sire”41, likely referring to Shire in western Tigray. Additionally, the 1683 map of Abyssinia by Ludolf and Gorgoryos places a region called “Sire” near Aksum, likely a variant of the term Shire.

This identification of Sigye as an early form of shire seems plausible, as Shire lies just west of Aksum and would have been a strategically significant region to conquer early on, due to its proximity.

Lastly, the conqueror claims to have seized half of the property (likely referring to livestock, goods, and people), likely to replenish his manpower and resources, thereby enabling his further campaigns.

Further Conquests: Aua, Tiamo, and Gambela

We now move on to the 2nd section of his conquests. The first people or territory he mentions conquering is a place called Aua. This might refer to the city of Adwa and the people who inhabited that region at the time. According to Henry Salt42, the name Adwa might derive from Adi-Awa (meaning land of Awa). Adwa is a city located northeast of Aksum in present-day Ethiopia.

The difficulty with this identification lies in the fact that the city of Aua was also mentioned by a Roman ambassador in the 6th century during a visit to Emperor Kaleb: ‘He tells us that Adulis is fifteen days’ journey from Axumis. On his way there, he and his companions saw a remarkable sight in the neighbourhood of Aue (Ave), midway between Axumis and Adulis’43. If Aua was indeed midway between Adulis and Aksum, this rules out Adwa, which lies barely 50 km from Aksum. This would instead suggest that Aua was located near Matara. However, it is also possible that these references are to two different places named Aua.

Next is Tiamo (or Tziamo). The exact location of this region or people remains uncertain. However, as with Sigye, the 11th-century copies of Cosmas’s Christian Topography include a footnote stating44:
“Τίαμα τοῦς λεγουμένος Τζιαμῶ καὶ τοὺς Γαμβελὰ καὶ τὰ ἐγγὺς αὐτῶν λέγει ἔθνη τὰ πέραν τοῦ Νείλου.”
This translates as: "He says that Tiama refers to those called Tziamō and the Gambēla, and the tribes near them across the Nile."

There are two ways to interpret the location of Tziamo. Firstly, the following section mentions the people known as the Semenoi, who are likely to have lived in the Sémen Mountains, situated further south of Tigray. This suggests that the regions previously listed, including Tziamo, likely lay in the general direction south of Aksum. Secondly, a 17th-century map by Ludolf and Abba Gorgoryos places Tzama directly to the south of Aksum, supporting this interpretation.

As for the Gambēla, they were possibly located in the same general area. The 18th-century traveller James Bruce mentions them as residing in the Enderta region of south-eastern Tigray45.

Conquests around the Sémen Mountains

As for the Zingabene, their precise location remains unknown. Our only clue comes from Henry Salt46, who associated them with the Bugna region, historically part of north-western Wollo province. Interestingly, Bugna was also the seat of the Begwenā Dynasty, commonly known as the Zagwe Dynasty. The area lies near the Täkkäze River, a tributary of the Nile.

The mention of the Athagaws next, which is almost certainly the Agaws, who historically lived around the Bugna area, further adds credence to this theory.

Regarding Angabe, we can likewise only speculate. Salt mentions a place called Angabet (or Andabet) in Begemder47, near the Gondar region. Kalaa, however, remains entirely unidentified. Salt does associate it with a location called Salat, which Ludolf’s map places in the Sémen Mountains. This would be consistent with the next people listed in the inscription, the Semenoi, clearly a reference to the inhabitants of the Sémen Mountains.

Interestingly, the Beta Israel (who lived in the Semien region) are referred to as “Kayla” (similar to Kaala) in some medieval Solomonic era manuscripts, particularly during the Gondarian period, although this may be a later attribution48.

The phrase “beyond the Nile” likely refers to crossing one of the branches of the Täkkäze River to get to the Sémen Mountains. The mention of frost and deep snow supports the conclusion that this campaign took place in the Sémen Mountains. Although snowfall there today is minimal, in the 2nd century AD, during the time of these conquests, the climate was likely colder, explaining the description of ‘snow reaching a man’s knees’.

Conquests Of The North - Beja & surrounding territories.

After conquering the Sémen Mountains, the conqueror and his army likely paused to rest, resupply, and consolidate their previous gains, given the challenging terrain and resistance from the local population. Also, the fact that the next series of conquests occurred much further north, beyond already subdued territories and even past Aksum, supports this theory.

The nations of Lazine, Zaa, and Gabala remain obscure to us. However, an educated guess can be made based on two key details from the inscription. First, it describes these people as inhabiting mountainous regions with steep declivities (κατάρρυτος – meaning sloping or flowing downward49), suggesting a sharp descent towards the lowlands. Second, the areas are noted for their hot springs and are stated before the mention of the Beja people; therefore, they likely lie near the Beja people (referred to as "Bega" in the text).

These clues point to a region in central to north-eastern Eritrea, likely within the Senhit and Sahel provinces (Parts of modern-day Semienawi Kayih Bahri & Anseba). For example, the southern Senhit area contains hot springs such as Maiwooi50, which align well with the inscription’s description. It is therefore quite plausible that the peoples of Lazine, Zaa, and Gabala once lived in this region, extending northwards towards the Atalmo and Beja territories.

Scholar Sergew Hable Selassie further suggests that Lazine, Zaa, and Gabala correspond to areas inhabited today by the Mensa (Tigrē) and Habab (Tigrē) tribes51, also known as the Habab region.

In the chronicles of Amda Seyon Conquests, the Gabal are located in the north-eastern part of Endarta in Tigray. G. W. B. Huntingford further observes that “Gabala may be Kabalé (Raheita)”52, most likely referring to a location near Rehayto, a town in the Southern Red Sea region of modern Eritrea, near the Djibouti border.

The Atalmo nation remains unidentified, but it likely lay just south of the Beja territories, as it is mentioned preceding them. The "Bega" is undoubtedly a reference to the Beja, who were also known as the Blemmyes in antiquity. They inhabited a broad region along the Red Sea coast in what is now Sudan, extending as far north as Berenike in Egypt53. The Tangattae were a Beja tribe54.

The final sentence highlights one of the main motivations behind the northern campaigns: securing a direct corridor to Egypt. As explored in detail in my article on Adulis, Egypt was one of the region's key trading partners of the highlands of Eritrea & Northern Ethiopia during antiquity. Ensuring access to this trade route was of immense economic importance and helped finance further military expansion.

Once again, the 11th-century copies of Cosmas’s manuscript provide marginal notes offering additional insight into this particular section of the inscription:

οὓς Βλέμμυας οὕτως καλοῦσιν οἱ Αἰθίοπες· τοὺς Ταγγαΐτας καλεῖ τοὺς Ἀτταβίτη καὶ τοὺς Ἀδρα
→ “Thus the Aethiopians call the Blemmyes. He refers to the Tangaitai as the Attabitae and the Adra”55.

This footnote further reinforces the identification of the Bega with the Beja people, as Blemmyes is the Greco-Roman term commonly used for the Beja during antiquity. The Attabitae and Adra nations are unknown.

Conquests Of Barbaria

The nation of Annine remains unidentified by modern scholars. The only detail provided is that they inhabited (δυσβατώτατον - hard to traverse/precipitous56) mountains, suggesting a rugged, mountainous terrain with cliffs adjacent to the lowlands. Since the subsequent conquests described in the text occur in the northern Somaliland and Djibouti region, these tribes must have lived further north, placing them possibly in the Dobe’a region of medieval Abyssinia (around the present-day Afar and south-east Tigray area).

The Metine are mentioned in an Aksumite-era inscription in connection with a place called HMS, which may represent an unvocalised form of Hamasen, a medieval province in the northern Eritrean highlands57.

Next, he conquered the Sesea nation, which refers to tribes located within Barbaria ( modern-day northern Somaliland). This is confirmed by the 11th-century footnote:
"Σέσεα ἔθνος ἐπολέμησα. Τὰ τῆς Βαρβαρίας ἔθνη ἐνταῦθα δηλοί."
→ "I waged war against the Seasea nation. Here, he clearly refers to the tribes of Barbaria”58.

There may be an etymological link between the Sesea and the Issa clan of Somalia59, who continue to inhabit northern Somaliland to this day.

Among the first of these tribes he encountered was the Rhausi, one of the inland peoples of Barbaria. They dwelled in arid plains, likely referring to the desert lowlands of the frankincense country (interior Somaliland). Controlling these tribes would have had significant economic importance, as frankincense was one of the most valuable trade commodities of the ancient world.

After his inland campaigns, the emperor turned back toward the coast and subdued Solate, a polity possibly located along the Somaliland coastline.

This section of the inscription also offers rare insight into the emperor's military strategies. For instance, he is said to have blockaded enemy tribes who had taken refuge in the mountains, starving them into surrendering. He also integrated the finest soldiers from conquered peoples into his own forces, simultaneously strengthening his military in addition to beginning the process of assimilating new groups into the empire.

Finally, the mention of imposing tributary status, rather than direct rule, suggests that local power structures were not entirely dismantled; rather, they were redirected to serve the emperor's authority, as was the case with the Solate.

Conquests Of Arabia

The final section of the Monumentum Adulitanum II describes the conqueror’s campaigns in Arabia, specifically on the coastline of present-day Saudi Arabia.

A combined fleet and land force was required to cross the Erythraean Sea, with the ships likely based at Gabaza, the port of Adulis. The inscription records the conquest of two Arabian tribes: the Arabitae and the Cinaedocolpitae, whose locations are identified through Ptolemy’s Geography and explanatory notes preserved in 8th- and 11th-century manuscript copies of Cosmas Indicopleustes’ book.

According to Ptolemy60, the Cinaedocolpitae occupied the north-central Arabian coastline (specifically 90km north of present-day Mecca61), while the Arabitae formed part of Arabia Felix, in what is now western central Arabia. In other words, much of the Tihāmah region north of Saba was likely brought under the control of the Aksumites, which is a view supported by scholars such as Sergew Hable Selassie62, I︠U︡riĭ Mikhailovich Kobishchanov63, and G. W. Bowersock64.

This theory is reinforced by the 8th-century manuscript (Vat. gr. 699), which has a footnote that states65: “Ἀρραβίτας καὶ Κιναιδοκολπίτας τοὺς εἰς τὸν Ὁμηρίτην σημαίνει, τουτέστι τοὺς ἐν τῇ Εὐδαίμονι Ἀρραβίᾳ” (“By ‘Arabitae’ and ‘Cinaedocolpitae’ he means those in the land of the Homerites, that is, those in Arabia Felix”). Likewise, the 11th-century manuscript includes a footnote that states66: “Here, he calls the Homerites ‘Arabites’, and the coastal Adonites he calls ‘Cinaedocolpitae’”.

Interestingly, cave paintings featuring scripts resembling Ge’ez have been discovered in north-eastern Saudi Arabia, such as the inscriptions at Tayma. This may indicate a possible presence of Aksumite or pre-Aksumite peoples in the general area.

The inscription also references Leuce Come, a well-known Nabataean port situated at the northernmost tip of the Red Sea (The Nabataean kingdom was conquered by Rome in 106 AD)67. Therefore, the conquests attributed to the Emperor imply that all territory between Roman Leuce Come and Saba became tributary to him. As a result, all trade passing through these lands would have been taxed, elevating his empire to the dominant maritime power in the Erthraean Sea and giving it control over the majority of regional trade.

Alliance with the Roman Empire?

Scholar Mikhail D. Bukharin68 argues that one of the key motives for the invasion of the South Arabian coastline was to aid the Roman Empire in suppressing disruptions to Red Sea trade. He cites a passage from the Periplus of the Erythraean Sea describing a group called the Carnaites (who he identifies with the Cinaedocolpitae) who are described as engaging in raids, kidnapping, and the enslavement of merchants.:

The final section of the inscription reaffirms the extent of his empire, specifically noting that it stretched westward, bordering both Ethiopia and Sasu. Eastwards it reached the `country of frankincense` (Barbaria /Somaliland).

In the geographical understanding of the time, “Ethiopia” referred to regions encompassing Sudan and parts of present-day Eritrea and Ethiopia. In the context of this inscription, it likely refers to the areas around Meroë, while “Sasu” likely lay further south, in territories corresponding to modern central and southern Ethiopia69.

It also records that the conqueror descended to Adulis, where, following his successful campaigns, he assembled his army. There, he commissioned the construction of a throne beside the stela commemorating the conquests of Ptolemy III, thereby linking his own achievements to those of the Ptolemies and, by extension, to Alexander the Great, founder of the Ptolemaic dynasty.

The closing lines of the inscription convey thanks to the gods Ares, Zeus, and Poseidon. In Aksumite tradition, these deities would be locally associated with Mahrem (Ares, god of war70), Baher (Poseidon, god of the sea71), and Astar (Zeus, a sky deity72).

Conclusion

So as we move into the early third century AD, the South Arabian inscription Ja 56773 records the name of Emperor GDRT, ruler of the Ḥabashat and the Aksumites. He would be the first in a line of emperors in the third century to intervene in South Arabian affairs. That, however, is a story for another time. For now, we can see that the conquests described in the Monumentum Adulitanum II were essential in uniting the highlands of Eritrea and Ethiopia, laying the foundation for the Aksumite Empire’s eventual rise in influence across much of the Erythraean Sea.

Further Readings

• Cosmas Indicopleustes, Christian Topography (1897) pp. 23-90. Book 2, section 142

• The Throne of Adulis: Red Sea Wars on the Eve of Islam, Chapter 4.

• https://pietroliuzzo.github.io/DHEth/rie277table.html → Good for cross-referencing place names

• The Historical Geography of Ethiopia from the First Century AD, pg 40-47

Subscribe now

Date Created: 6/20/2025

Last Updated: 1/3/2026

Adulis (Greek: Ἄδουλις) breaks down as follows:

• Ἀ- (a-) → a prefix meaning “without”

• δουλ (doul) → from δοῦλος (doûlos), meaning “slave” (masculine)

• -ις (-is) → a common noun suffix in Greek

Altogether, Ἄδουλις has been interpreted by some scholars as meaning “place without slaves.”
However, this is one of several proposed etymologies and remains debated among historians and linguists1.

---

Adulis (Saho: ado lay) can be broken down as follows:

• ado → ‘white’

• lay → ‘water’

Referring to clear water, this term is used by the Saho people to describe various settlements known for having access to clean water sources2.

---

Adoulîtai (Greek: Ἀδουλῖται) breaks down as follows:

• Ἀδουλ- → root from Ἄδουλις (Adulis)

• -ῖται (-îtai) → a plural suffix denoting, people of a place

Altogether, Ἀδουλῖται means “the people of Adulis”

This is the form commonly used in Greek manuscripts, including in Ptolemy's Geography (Book II–VI, 4.7.27)/Greek Version

---

Adoulîtē (Greek: Ἀδουλῖτης ) breaks down as follows:

• Ἀδουλ- → root from Ἄδουλις (Adulis)

• -ῖτης (-îtēs) → a masculine suffix indicating “a person from a place”

This form used by Stephanus of Byzantium, in his work ‘Ethnica’.

Introduction

Adulis was one of the most important trading cities on the Erythraean Sea during antiquity. Merchants and travellers from as far as Rome, India & China visited the shores of the Gulf of Zula, where Adulis’s port, Gabaza, was situated. At one point, during the reign of Emperor ካሌብ (Kaleb) in the early 6th century AD, the port hosted a fleet of over 60 ships and tens of thousands of soldiers, serving as the launch point for Kaleb’s military expedition into Himyar. Around the same time, Adulite merchants were recorded by Cosmas Indicopleustes as crossing the Bab al-Mandeb and traversing the Persian Gulf, reaching as far as Sri Lanka3. Mercentile activities were just one domain, as Adulite priests were recorded travelling to the courts of Constantinople to participate in the vibrant theological councils and religious debates of the 5th century4.

A contentious topic of discussion is whether Adulis was merely a port city under Aksumite rule or an entirely independent entity. The reality lies somewhere in between. As we will soon explore, Adulis predates the rise of the Aksumite Empire as it existed in some form as far back as the Puntite era. Therefore, it cannot be described as just an Aksumite port. However, during the later phases of Adulis’s history (particularly between the 3rd and 6th centuries AD), it was, at least during certain periods, incorporated into Aksumite control. This is evident in the writings of travellers such as Cosmas Indicopleustes in the 6th century AD, who refer to it in as an Aksumite port . Still, even during this period, Greek sources often distinguish between Adulites and Aksumites, suggesting that full integration was never entirely realised. That said, this article focuses on the period prior to the 2nd century AD, before the emergence of the Aksumite Empire, so a deeper discussion of Adulis’s later relationship with Aksum will be reserved for a subsequent article.

1st Millennium BC & Before

Nevertheless, the history of Adulis extends far further back than the 4th and 5th centuries AD. Recent archaeological evidence, including obsidian samples5 and genetic tracing from baboons6, links Adulis to the ancient Land of Punt (as discussed in greater detail in my article on the Land Of Punt), this suggests that Adulis may have been one of its key ports, thereby potentially dating the port cities origins as far back as the 2nd to 3rd millennium BC.

Unfortunately, little is known about the period following the Land of Punt (~1000 BC onwards) up to the Ptolemaic Era of Egypt (~4th–3rd century BC). This gap is largely due to the limited archaeological research conducted at Adulis. However, other reasons for the lack of information might include the decline in trade with Egypt (and by extension, with other Mediterranean civilisations) during the mid to late first millennium BC. This downturn coincided with the weakening of the Egyptian state during the Third Intermediate Period, as discussed in my section The Fall of DʿMT in the DʿMT article. Trade would later be revived throughout the Erythraean Sea during the reign of the Ptolemaic dynasty, beginning around the 4th century BC.

Archaeological work at Adulis is ongoing, and just last year, new evidence of settlement, such as ceramics and other pottery dating from 1500 BC to 100 BC was uncovered7. These findings may offer valuable look into this little-known era of Adulis's early history.

This article focuses on the earlier history of Adulis, specifically from around 300 BC to 200 AD, a period of around half a millennium. To reconstruct this period, I will draw upon both primary and secondary sources. Many of the primary accounts come from contemporary Greek writers, who referred to the inhabitants as Adoulîtai /Adulites and the city as Adule or Adulis. As we will see, cultural ties with the Greek world were particularly strong during this time. Given that Adulis was a merchant city during an era when Greek served as the lingua franca of regional trade, this is hardly surprising.

In addition, primary inscriptions and artefacts discovered at Adulis itself will also be examined to give a better understanding of what occurred during this period.

Location Of Adulis

Adulis, now located over 5 kilometres from the Gulf of Zula, was theorised to have been much closer to the sea in ancient times, approximately 3.3 kilometres in fact8. Due to siltation, this ancient city now lies much further away from the coast. Adulis is situated next to three rivers, with the northernmost one being the Haddas River, on which Adulis is located. Merchants used the Haddas River to help guide them on their travels southwards to the town of Qohaito and further south all the way to Aksum and beyond….9.

The port of Gabaza

While Adulis served as the bustling port city where merchants conducted trade and local rulers resided, the actual harbour where ships moored was likely at a site called Gabaza, as mentioned in the text Martyrs of Najran, which describes the 6th-century AD military conquest of the Himyarite kingdom by Emperor Kaleb of Aksum.

It reads as follows:

Archaeologists have recently proposed that Gabaza was located in the Galala Hills, south of Adulis, where the ruins of two ancient circular watchtowers and pottery resembling the region’s antiquity style have been discovered10.

Gabaza was also mentioned in an Aksumite inscription (DAE 811) , scholars suggest this is possibly in reference to a King of the port town12.

Samidi

Samidi, a town mentioned on Cosmas Indicopleustes’ map, may correspond to a site located 7 km north of Adulis. Several ancient architectural fragments have been discovered there, along with a mound containing the ruins of what may have been a large building measuring 40 metres in length and 25 metres in width13.

The Great Stele of Adulis (~241- 221 BC)

Two of the most significant monuments once found at Adulis—now lost to history—were the Basilic Stele of Adulis and the Marble Throne. Each commemorated the conquests of different rulers from different eras: the stele recorded the campaigns of Ptolemy III, while the throne belonged to an unnamed indigenous ruler from the Common Era onwards. Both held immense importance for the Adulite people. As noted by Cosmas Indicopleustes, criminals were even executed in front of the throne & Emperor Kaleb made it a priority to copy these inscriptions before heading out for his expedition (It’s no doubt important to learn the conquests of others before attempting one yourself).

Therefore, to understand the earliest part of Adulis’ history, it’s important to understand the contents of this Basilt Stele.

The Structure Of The Stele

When the 6th century Explorer/Merchant Cosmas Indicopleustes ventured to Adulis at around 522 to 523 AD 14 , he came as part of a group of merchants15 sent forth from Alexandria in Egypt, when he landed at Adulis he remarked at the marble throne that he had observed, this throne had inscriptions written on the sides, in addition to the throne lay a stele made of basalt. Cosmas notes that the Emperor ካሌብ/Kaleb (Greek: Ἐλεσβαὰν → Elesbaan) had asked the Ruler (ἄρχων→ắrkhōn) of Adulis (whose name is ᾿Ασβᾶς/Asbas/Abbas) to have a copy of the inscriptions written down and sent to him. Abbas then assigned this task to Cosmas and his fellow Merchant Menas, knowing both were proficient in Greek and could thus decipher the Greek written on the stele & throne.

Κοσμᾶς Ἰνδικοπλεύστης → Cosmas (Universe/Order) + Indikós(Indian) + pleustēs (Sailor).

Luckily for us, Cosmas and Menas copied the inscription more than once, with one copy being given to Abbas, while the other was kept for themselves.

The following excerpt regarding the throne and stele is taken from Cosmas Indicopleustes:

Cosmas mistakenly attributes both the Stele and the Throne as being created by the same person and at the same time. However, in reality, scholars believe the Stele was likely erected much earlier than the Throne. The Stele describes the conquests of Ptolemy III, while the inscriptions on the Throne detail the conquests of an indigenous ruler of the land16.

It then continues…

Hercules was the Roman equivalent of the Greek legendary hero Heracles, the son of Zeus and the mortal woman Alcmene, while Mercury corresponded to the Greek god Hermes, the god of merchants. Could this allude to the importance of War & Trade by the inhabitants of Adulis?

In both excerpts, we find information regarding the structural details of the Stele and the Throne. Focusing on the Stele, the Stele is said to be made of Basalt. Basalt is a common type of volcanic rock found in the Danakil Depression, south of Adulis, so it is plausible that these rocks were mined and used to construct the Stele17.

Secondly, rectangular steles have been discovered throughout the highlands of Eritrea and Ethiopia, dating back to the early and mid-first millennium BC. For example, several such steles have been found at Käskäse (as discussed in greater detail in my article on Early Antiquity Societies in Eritrea’s Akkälä Guzay region). This suggests that the stele in question was very likely crafted indigenously. We can also ascertain that the stele was around 1.5 meters in height (one ancient Greek cubit is around ~0.4-5 meters in height18).

From the above excerpts, we also understand that by the time Cosmas visited Adulis in the early 6th century, the stele had been broken, with its lower part destroyed. While the exact reason for this remains uncertain, it is highly unlikely that it was left unrepaired due to a lack of will. It is more plausible that the stele was deliberately broken and left in that state because it was disfavored by the local inhabitants of Adulis and the regional king, particularly due to the inscriptions on its lower part. However, we can’t be sure until more evidence is brought to light.

So, what exactly was inscribed on this basalt stele, and when was it created? As we shall see, the stele records the military campaigns of Ptolemy III Euergetes, who reigned between 246 and 222 BCE. But this raises an intriguing question—why would a monument commemorating him be located thousands of kilometres away from Alexandria? Let’s find out…

The Ptolemaic Era(~305BC-30BC)

The Ptolemaic Dynasty was the last dynasty that ruled Egypt prior to its annexation by Rome. It was founded by Ptolemy I, a general of Alexander the Great and it lasted roughly 300 years (305BC to 30BC)19.

As previously discussed, the Ptolemaic dynasty ushered in a new era of trade in the Erythraean Sea, particularly during the reigns of Ptolemy II Philadelphus and Ptolemy III Euergetes. Ptolemy II established several coastal ports along the Erythraean Sea, including Philotera and Ptolemais Theron (“Ptolemais of the Hunts”)20, the latter named in honour of his fondness for hunting in the region.

Ptolemy II’s passion for hunting was not merely recreational. Elephants played a crucial role in warfare (nicknamed the "Tanks” of antiquity), particularly in conflicts against their chief rivals, the Seleucids, who ruled over Persia, Mesopotamia, the Levant, and parts of Anatolia (modern-day Turkey). The seleucids restricted access to Indian elephants, thus it became an imperative for Ptolemy II to secure an alternative source. As we shall see, that source lay in the region surrounding Adulis…

It is not difficult to imagine that the expansion of trade and the increased maritime security provided by Ptolemaic Egypt contributed to Adulis’s revival following the economic downturn of the preceding era.

The extent of Ptolemy II power can’t be understated: “[Ptolemy] takes slices of Phoenicia and Arabia and Syria and Libya and the dark-skinned Ethiopians; all the Pamphylians and the warriors of Cilicia he commands, and the Lycians and the Carians, who delight in war, and the islands of the Cyclades, for his are the finest ships sailing the ocean.” - (Theocritus, lines 90-91)

The Inscription On The Stele

Ptolemy II’s son, Ptolemy III, would then utilise the newly captured elephants in his wars against the Seleucids. These wars are described in the Stele of Adulis as follows:

Analysis

The conquests mentioned in the inscription refer to the Third Syrian War21, which was fought between the Ptolemaic Dynasty and the Seleucid Empire between 246 and 241 AD22. The conflict was sparked by a succession dispute following the death of Antiochos II. His wife Laodike claimed the throne for her son, Seleukos II, asserting that Antiochos II had named him heir at his deathbed. However, Berenice Syra (daughter of Ptolemy II, sister of Ptolemy III, and also wife of Antiochos II) contested this claim, insisting her own son was the rightful successor. She appealed to her brother Ptolemy III for support, but before action could be taken, both she and her son were assassinated. This event prompted Ptolemy III to declare war on the Seleucid Empire23.

Interestingly, he notes the use of war elephants captured from regions referred to as Troglodytes and Ethiopia, both fluid geographical toponyms in antiquity. During the time of Ptolemy III, "Troglodytes" generally denoted the area stretching from just north of Adulis to southern Egypt (around Berenice), primarily inhabited by the Beja people. "Ethiopia," on the other hand, referred to a broader region encompassing parts of present-day Sudan, Eritrea, and Ethiopia.

Troglodytes (Greek: τρωγλοδύτης24) breaks down as follows:

• τρώγλη (trōglē) → “hole”

• δύω (dúō) → “to enter” or “to dive into”

• -της (-tēs) → a suffix indicating a person/people in this context.

Altogether, τρωγλοδύτης means “one who creeps into holes” or “cave-dweller”.

---

Aithíops (Greek: Αἰθίοψ25) breaks down as follows:

• αἴθω (aíthō) – “to burn”

• ὤψ (ṓps) – “face” or “appearance”

So Αἰθίοψ means “burnt-face”, a term used by ancient Greeks to describe people with a dark skin complextion, typically those living in the regions of present-day Sudan, Ethiopia, and Eritrea.

Ptolemy III then claims to have employed these war elephants to capture territories across the Levant up to the Euphrates River in what is now modern-day Iraq. From there, he reportedly advanced northwards into Anatolia (present-day Türkiye). While many of these conquests are attested26, their extent is likely exaggerated.

He further proclaims to have crossed the Euphrates, subdued Mesopotamia, and pushed into Persia, claiming victories as far as Bactriana (a region encompassing parts of present-day Afghanistan, Tajikistan, Uzbekistan, and Turkmenistan). However, scholars question whether he ever held direct control over these distant regions, or even truly conquered them27. It is more plausible that the term “subdued,” as used in the inscription, implies forcing these territories into a state of peace (which did occur in 261 BC28).

The final section of the inscription notes that a small portion of the lower part of the inscription was fractured and presumably lost. As previously discussed, this may have been the result of deliberate defacement rather than environmental or accidental damage. One must wonder what was written there to provoke such an act. In many cases, the closing lines of monumental inscriptions are dedicatory, offering praise to the gods for the ruler’s triumphs.

Nevertheless, one thing is clear: war elephants played a crucial role in Ptolemy III's military campaigns. Since these elephants were sourced from the region around Adulis, it is unsurprising that a stele was commissioned and erected there, likely as an expression of gratitude and to enhance the ruler’s prestige in one of the farthest reaches of the known world at the time - Adulis.

Battle Of Raphia

Scholars have also proposed that the African elephants used by Ptolemy IV, in the Battle of Raphia in 217 BC, were sourced from Adulis and its surrounding region29. According to the Greek historian Polybius, approximately 73 African war elephants participated in the battle.

However, scholars remain divided on whether these were savanna (larger) or forest (smaller) elephants. The ancient historian Polybius described the African elephants as smaller than Asian elephants, but the accuracy of this claim is hotly debated30. While Eritrea's modern elephant population consists of the larger savanna type, some scholars suggest that the ancient population may have differed from today’s.

Elephants In Eritrea

In modern times, Eritrea is home to one of the northernmost populations of Elephants in Africa31. Today, only around a hundred elephants remain in Eritrea, primarily confined to the Gash-Barka region. However, this was not always the case. In the 6th century, the Roman ambassador Photius, who travelled from the port of Adulis to Aksum, reported encountering a large herd of elephants roughly halfway along his journey (likely near present-day Matara).

The use of elephants in warfare by the Ptolemies strongly implies that they were trained for combat beforehand. It is likely that this training occurred partly, if not entirely, in the region around Adulis before the elephants were exported to Egypt by merchants32.

Elephants must be reared and trained from a young age, particularly for warfare. The process starts by fostering a relationship with older elephants; through interactions such as play-fighting, the younger elephants then come to recognise the trainer as part of the social hierarchy and form bonds accordingly33.

The mahout—the rider and handler of the elephant—relies on a specialised technique involving a metal rod used to tap the elephant’s ears. The location and intensity of these taps communicate different commands34. Notably, two bronze mahout rods have been discovered at Adulis, further supporting the idea that elephant training took place there.

48 BC ~ 77 AD

Pliny The Elder

Although Pliny the Elder composed his works around 77 AD, much of his information about Adulis came from the earlier writings of King Juba II of Mauretania, who lived from around 48 BCE to 23 AD35. This is further evidence of Adulis's significance as a port city during the 1st century BC.

Pliny the Elder’s Natural History (Latin: Naturalis Historia), published around AD 77 and hailed as the first encyclopaedia despite its imperfections, contains our earliest surviving reference to Adulis. In Book VI, which deals with geography, he mentions Adulis several times, for instance, the following excerpt details the location of Adulis:

Some scholars have hypothesised that "Daphnidis" may have referred to the Dahlak Islands.

Adulitae → Latin: Aduliton

Origins of Adulis, Freeman’s town?

Firstly, Pliny mentions a legend suggesting that Adulis was founded by Egyptian slaves. This tale might be connected to the city's name, as 'douloi' means slaves in Greek, and the prefix 'a-' indicates a lack, thus implying 'a lack of slaves'36. Although there is no corroborating evidence for this etymology.

Nowdays, this is similar to “Freemans” town.

The story may refer to runaway slaves who arrived in Adulis during the earlier Ptolemaic period. Alternatively, it could trace back to the much older ties between Egypt and the land of Punt, a relationship that spanned millennia (as noted in my article on Punt). Given this long-standing connection, Pliny’s association of Adulis with former slaves becomes more understandable. Egyptian slaves, perhaps employed by merchants as labourers aboard ships, may have escaped and settled among the local population, leading to the tradition he records.

A Glimpse Into Trade

The excerpt also highlights just how significant Adulis was as a port in the Erythraean Sea. It served as the primary maritime hub for both the Troglodytæ and the peoples of Æthiopia, regions that, as discussed earlier, stretched from Egypt deep into the Horn of Africa. This suggests that a substantial volume of trade passed through Adulis. Pliny notes that Adulis lay five days’ sail from Ptolemais, likely referring to Ptolemais Theron, a trading port near present-day Aqiq in Sudan.

He lists several exports from Adulis, chief among them being ivory, which he notes was shipped in large quantities, suggesting it was the region's primary export. Other goods include rhinoceros horn, hippopotamus hides, and tortoise shell, likely sourced from the nearby Dahlak Islands. Pliny also mentions "sphingia," a term that appears to be the plural of sphingion37, possibly referring to a type of ape. This could correspond to Papio hamadryas, the same species exported from Punt to Egypt in earlier periods. Finally, he notes the export of slaves, likely prisoners from local conquests, which was a common practice during that era.

Maritime Trade

Trade route with Egypt

Pliny also describes how “Ethiopians” engaged in maritime trade with ancient Egyptians. For example, in his writings, he mentions that at the Island of Elephantine, located in Upper Egypt, the Ethiopians would leave their boats and carry them to avoid the rocks at each cataract.

Ethiopians as described by Pliny The Elder, described a people group/ “race” that lived South of Egypt, in present day Sudan, Eritrea, Ethiopia , the largest city and proposed capital was Meroe

“Again in Meroe — this is an inhabited island in the river Nile 5000 stades from Syene, and is the capital of the Aethiopian race” -Pliny The Elder “Natural History", Book 5, Chapter 2, Section 184

As previously mentioned, the term Aethiopian was used as a broad label for dark-skinned peoples south of Egypt. At the time, Meroe was the dominant power and thus became the primary reference for Aethiopia. Similarly, following the fall of Meroe to the Aksumite Empire in the late 3rd century AD, the Aksumite Empire and its surrounding region gradually came to be identified as "Ethiopia" by the Greco-Romans foreigners.

This is akin to using the word “North East-African” today.

It is highly likely that trade with Egypt also took place via an inland route, utilising the Nile’s tributaries and navigating through its cataracts. This route was probably established as far back as the Puntite period. The native inhabitants of the region — whether from Meroe, the Troglodytae, Adulis, or Aksum — likely used this inland corridor alongside the maritime routes of the Erythraean Sea.

Early Adulis Boats

Pliny also describes some of the vessels used for maritime activities. Specifically, he details how cinnamon was acquired by the Aethiopians and then traded across the sea using simple boats that lacked sails or oars, relying instead on the natural winds to guide them across the Arabian seas. In return, they received goods such as glass and copper, and highly demanded items like necklaces, bracelets etc. The excerpt reads as follows:

Although Adulis had its own inhabitants, known as the Adulitae or Adulites, its role as a merchant city attracted a diverse range of peoples. Traders from across north-east Africa gathered there, along with Egyptians, Romans, South Arabians, and Persians etc...

Gebbanitae were likely a group in Southern Arabia, next to the Sabaeans and Hydramiates38. Ocelis is located near the Bāb al‑Mandab strait and was frequently used to set sail onwards to India39. The fact that it takes nearly five years for merchants from Adulis to return suggests they were travelling vast distances, possibly as far as India. These merchants returned with glass and copper, clothing, and buckles, bracelets and necklaces. As we will see later, the Periplus Of The Erythraean Sea will also mention these same imports.

As mentioned earlier, by around 500 AD, Cosmas confirms that merchants from Adulis were present in Sri Lanka. Therefore, it is not far-fetched to consider that Adulite merchants may have been undertaking such extensive journeys even earlier.

“Consequently, that traffic depends principally on having the confidence of the women" — this may suggest that long-distance trade by the Adulites was, in part, driven by the desires of local women for ornaments such as bracelets and necklaces. Over time, such items became integral to women's dress in the region. To this day, jewellery, necklaces, bracelets, earrings, and more remain an essential part of women's attire and cultural expression in the region…

Markets In Adulis

Dr Daniel Habtemichaels website is excepetional for more Information on Adulis.

So, where did merchants unload their textiles, raw materials, and other goods after docking at the port of Gabaza? And where did traders from Adulis, Aksum, and neighbouring regions bring their goods for exchange?

Much of this commercial activity likely took place in and around the markets of Adulis. One such market has been identified through archaeological excavations led by Dr Daniel Habtemichael. The structure was made from stone and measured approximately 18.12 by 15.40 metres, with remnants of roof-supporting pillars and a lower storage level for goods40. Multiple examples of this market type have been uncovered at Adulis.

These structures differ from those in places like Qohaito in their size. While Qohaito's markets typically featured two-pillared designs41, those at Adulis had four, reflecting the higher trade volume that Adulis handled as a major Red Sea port.

Archaeologists estimate that about 70 percent of all ancient coins discovered in the northern Horn of Africa have been unearthed at Adulis42.

Periplus of the Erythraean Sea (~50AD-~100AD)

Περίπλους τῆς Ἐρυθρᾶς Θαλάσσης (Periplous tēs Erythrâs Thalássēs) breaks down as follows:

• Περίπλους (Periplous) → from peri- (around) + -plous (sailing), meaning “a sailing around” or “voyage”

• τῆς (tēs) → “of the”

• Ἐρυθρᾶς (Erythrâs) → feminine form of the word “red”

• Θαλάσσης (Thalássēs) → feminine form of the word “sea”

Altogether, Περίπλους τῆς Ἐρυθρᾶς Θαλάσσης translates to “Voyage of the Red Sea”.

The term Periplous refers to a genre of ancient Greco-Roman texts that listed coastal ports, trade hubs, and general sailing directions. Helping sailors and merchants navigate foreign waters, the Periplus of the Erythraean Sea was one such document.

Modern-day Eritrea (Greek: Ἐρυθραία → Erythraía) derives its name from the Erythraean Sea, the ancient Greek term for the Red Sea.

The Periplus of the Erythraean Sea, whose author is unknown, is generally dated between 40 and 70AD43 (Scholar Wilfred Harvey Schoff proposed a more specific date of around 60 AD44). Despite the uncertainty surrounding its authorship, the text offers a detailed account of Adulis—its trade, surrounding territories, ruling authority, and the extent of that ruler’s domain. I will begin by presenting an excerpt concerning the trade and territorial context of Adulis:

“Below Ptolemais of the Hunts, at a distance of about three thousand stadia, there is Adulis, a port established by law, lying at the inner end of a bay that runs in toward the south. Before the harbor lies the so-called Mountain Island, about two hundred stadia sea-ward from the very head of the bay, with the shores of the mainland close to it on both sides. Ships bound for this port now anchor here because of attacks from the land. They used formerly to anchor at the very head of the bay, by an island called Diodorus, close to the shore, which could be reached on foot from the land; by which means the barbarous natives attacked the island. Opposite Mountain Island, on the mainland twenty stadia from shore, lies Adulis, a fair-sized village, from which there is a three-days' journey to Coloe, an inland town and the first market for ivory...” - Periplus of the Erythraean Sea, section 4

Ὀρεινὴ is the exact Greek word used in the Periplus, which Wilfred Harvey Schoff translated as "Mountain Island." It derives from the ancient Greek word ὄρος, meaning "mountain”.

Locations

As discussed earlier, Ptolemais of the Hunts was a port founded by Ptolemy II, located near the Gulf of ʿAqīq in present-day Sudan45. The Periplus later refers to Adulis as a port “established by law”, indicating that it was under the control of a formal government or legitimate authority (we will see later who this authority likely was). It is described as lying deep within the gulf; however, this is not entirely accurate, as Adulis is situated much further north, as can be seen in the map above. This discrepancy may simply be a minor geographic error, or it could reflect environmental changes, such as coastal shifts from flooding, over the past two millennia.

Adulis was one of only three ports along the entire Erythraean Sea route designated as “established by law” (νόμιμον → comes from the root νόμος → nómos, meaning law46), meaning the safety of traders and merchants was guaranteed upon arrival47.

The phrase “Mountain Island” (Ὀρεινὴ) likely refers to Dese Island48, located just off the coast of the Buri Peninsula. Based on my rough measurements online, the distance from the head of the bay to this island is approximately 44 km, or around 200 stadia, which aligns with the figure mentioned in the Periplus. In contrast, the location of the “Island of Diodorus” isn’t entirely known; some scholars speculate that it may have disappeared due to siltation49, while archeologists David Peacock & Lucy Blue speculate it was west of the Galala Hills, as analysis found that the area could have been surrounded by sea & many fragments foreign & local pottery were found here50.

The author of the Periplus also notes that native groups attacked vessels docking at the head of the bay. This may suggest that not all areas within the Gulf of Zula were governed by the same laws that applied to Adulis itself. Regions further from the port, such as the area around “Mountain Island”, may have been home to autonomous tribes or communities that operated independently of Adulis’ legal authority at certain periods.

Another possibility is that neighbouring tribes were permitted to act with hostility towards foreign merchants if existing trade agreements had expired/breached etc...

Inland trade

Next, the text reveals the inland trade route used by merchants from Adulis, which is described as a "κώμη σύμμετρος" (a fair-sized village). This route leads to Coloe (likely modern Qohayito), which is designated as a "πόλιν" (a form of "πόλις” - Polis, meaning city-state); the designation of town is erroneous. Scholars like J. W. McCrindle translated it as a city (see section 4 here). This classification for Coloe is logical, given its location in the more temperate highland climate, which would have supported more robust agriculture. Such conditions foster the development of larger state formations, supporting a greater population and a more complex society, hence a city. Coloe's advantageous position as the first highland destination from Adulis significantly contributed to its growth. The passage then states that it's a further five-day journey from Coloe to the "μετρόπολιν" (metropolis) of the Aksumites, implying it was their principal and largest city of the Aksumite people, likely referring to Aksum itself.

Exports

The text lists the region's main exports as elephants and rhinoceros, confirming Pliny's earlier accounts. Notably, the Periplus author mentions elephants could be found near Adulis - quite different from today, when elephants survive only in small numbers in Eritrea's western Gash-Barka region.

The author then describes many "sandy islands" called Alalaei, referring to the Dahlak Islands51. These islands were valued for their tortoise shells, which were also documented by Chinese historians during the Han Dynasty (as we will see later…).

Finally, "Fish-eaters" (Ichthyophagoi) was a Greek-Roman term for various coastal peoples whose diet was mainly fish. In the Periplus, it refers to an unknown group living near the coast who hunted tortoise shells (possibly the Dahlak people???).

The King Of The Erythraean Sea: Zoskales (Ζωσκάλης)

The Periplus then provides a rare glimpse into one of the earliest known rulers in the region during antiquity (outside of DM`T): a man named Zoskales. Interestingly, his character is detailed as follows:

He is described as ἀκριβὴς (from ákros, meaning precise or sharp52). This term has been translated by scholars like J. W. McCrindle and Wilfred Harvey Schoff as "penurious" or "miserly" (stingy). This interpretation might stem from the Greek word δὲ(dé, meaning “but”53) being used later in the description, suggesting a contrasting or excessive quality to his precision, perhaps specifically in financial dealings, given Adulis's focus on trade.