A 4,000-year-old wall was found hiding underneath a Moroccan ridge that a French archaeologist sketched in the 1930s

Charcoal beneath a wall Armand Ruhlmann recorded in 1936 at Oued Beht, Morocco, dates its first phase to as early as 2010 BC.

In 1936, the French archaeologist Armand Ruhlmann published photographs and a sketch of the walls on a ridge at Oued Beht, in Morocco's Khémisset Province. He took them for a prehistoric enclosure. Under the stones he described, a wall dated about 4,000 years old remained buried.

It turned up in a trench dug by hand across the largest wall during the 2023 field season, which ran from October 30 to November 17. Charcoal from its cobble foundation gave a radiocarbon range of 2010 to 1747 BC.

The trench belongs to the Oued Beht Archaeological Project, with partners in Rabat, Rome and Cambridge. Its results are now out in Libyan Studies.

The paper reports a ditch, an earth bank and a wall from the early to mid-second millennium BC, and it leaves their purpose open.

Ruhlmann's barred spur

Four stone alignments run east to west over the southern end of the ridge, above a river that flows all year. The village of Aït Siberne shows to the east in drone photographs taken by Toby Wilkinson of Cambridge.

Ruhlmann tied the biggest alignment, now labeled Wall B1, to a slighter wall on the western scarp and to Wall C further south. He read them as an éperon barré, or barred spur. That is a headland closed off on its open side by a rampart.

Giulio Lucarini of Italy's National Research Council and Cyprian Broodbank of Cambridge are the corresponding authors of the new paper. They and 19 co-authors call Ruhlmann's plan a "neat but unproven reconstruction."

A French and Moroccan project dug into Wall B1 and Wall C between 2013 and 2015. That work is unpublished, and the remnants of its trenches were still open in 2022 and 2023.

Local residents told the team that the upper stonework had been bigger within living memory. The wall had also been turned into a stable for animals.

Charcoal beneath Wall B1

Trench 3 was laid out as a long slit across Wall B1. Its northern extension, Trench 3.1, was offset to keep clear of an olive tree. Work there stopped at about 4 feet (1.2 meters) for safety.

The visible wall, numbered F.300.2, is about 7.5 feet (2.3 meters) wide at ground level and 4.4 feet (1.35 meters) high. It is built of limestone blocks and cobbles in compacted earth.

Under it lay F.300.1, an older structure of burnt and unburnt daub, a building mud, with ashy lenses full of charcoal. It sat on rounded river or beach cobbles less than 4 inches (10 centimeters) across. All 17 sherds from it were prehistoric.

A second charcoal date, from the burnt material, gave 1896 to 1623 BC. The authors link that layer to the destruction of the first wall.

The team had expected Wall B1 to widen the site's time range. "This expectation was spectacularly fulfilled," the authors wrote.

Polished axe in the ditch

North of the wall, Trench 3.1 ran into a ditch parallel to it. It was about 6 feet (1.84 meters) wide and at least 2 feet (65 centimeters) deep, and its base may not have been reached. One of its fills held a polished stone axe, and all 18 sherds from the ditch were prehistoric.

Bulk soil samples of 10 to 20 liters (about 3 to 5 gallons) went through flotation, which lifts seeds and charcoal off in water. Three charred seeds of domesticated crops from the ditch came back at about 1500 to 1300 BC.

The lowest phase of the earth bank gave 1532 to 1424 BC and 1497 to 1300 BC. A slanted posthole north of the wall, probably for a timber prop, held a legume seed dated to 1502 to 1324 BC.

Altogether the two trenches produced 17 dates, eight on seeds and nine on charcoal. Two charcoal samples with modern results were treated as intrusive and set aside.

The wall line follows the contact between Jurassic dolostone on the south and the Blue Marl Formation on the north. In their conclusions, the authors call the Wall B1 sequence "the earliest such constructions confirmed in the region." Defense is one reading, but other motives for enclosure "cannot be excluded at this stage," they wrote.

Barley and axes on the northern plateau

An earlier paper in Antiquity dated a farming community on the northern ridge to about 3400 to 2900 BC. Its authors called it currently the earliest and largest agricultural complex in Africa beyond the Nile corridor.

After the 2023 survey, that settlement is put at 10 to 11 hectares (25 to 27 acres). It has yielded bell-shaped storage pits, painted pottery and polished stone axes. Plant remains include barley, durum wheat, pea and wild olive.

In the second millennium BC, the activity moved south to the narrow tip. Morocco is better known for far older material, such as shell beads 142,000 to 150,000 years old and hominin jawbones from Casablanca.

The paper sets Oued Beht next to Kach Kouch, a site near the Mediterranean coast. People lived there late in the second millennium BC.

The authors suggest the two may mark a Bronze Age in northwestern Morocco, in step with the Argaric culture, a Bronze Age society in southern Spain. In those centuries, a fire at Cabezo Redondo in Spain preserved a wooden loom.

Rebuilt in the 11th century

Medieval builders went back to the same line. Charcoal from the middle phase of the bank gave two dates in the 11th to 12th centuries AD. A stone wall just north of Wall B1 gave the 11th to 13th.

The standing stone wall belongs to that later phase, according to the paper, because the upper layers of the bank probably went up at about the same time. A seed from the fill inside the wall gave 1729 to 1516 BC, which the authors put down to older soil reused as fill.

At the northern edge of the site, local people had reported human remains in 2022, during house building. The team dug three test trenches there. Trench 7 exposed a burial with its skull to the west, and a molar dated it to the 13th century AD.

Across a shallow valley, about 1,640 feet (500 meters) away, survey walkers found a dense spread of 11th to 13th century pottery on a hilltop. One sherd, painted cobalt blue, was likely made in Fez.

Wall C has not been dated

Wall C, the southern piece of Ruhlmann's plan, has no dates. Whether the three walls ever formed one enclosure is still unknown.

A copper-based rod 2.2 inches (5.5 centimeters) long came from the bank south of Wall B1. Its layer has no radiocarbon date, and no Bronze Age metal has been confirmed at Oued Beht so far.

Each radiocarbon date gives the earliest possible age of its layer. Charcoal and seeds can predate the soil that buried them.

Ruhlmann's 1936 sketch showed Wall B1 as a rampart of unknown age. Its first phase now has two charcoal dates, and the authors have proposed limited work on Wall C.

The project runs under a memorandum of understanding between Morocco's National Institute of Archaeological Sciences and Heritage, the National Research Council of Italy, ISMEO (the International Association for Mediterranean and Oriental Studies) and Cambridge's McDonald Institute. The paper thanks the Oukhayi family for letting the team dig on their property.

The full study was published in the journal Libyan Studies.

-- -

Check us out on EarthSnap, a free app brought to you by Eric Ralls and Earth.com.