55 km east of AlUla, and across the north-west, Saudi Arabia

Mustatils

Late sixth millennium BC, on cattle bone (ritual courts, not stock pens, and not a shared sunrise)

An aerial view of a long narrow stone rectangle in sand, with labels for the head, walls, courtyard, entrance and base. — One mustatil from the air, structure IDIHA-F-0003301, with the paper’s labels. The head is the far platform, the long walls enclose a courtyard, and the base and entrance are at the near end. An I-type platform sits beside it. This is not the excavated monument whose bone was dated. Kennedy and co-authors, CC BY-SA 4.0. Shown unedited.

From the ground a mustatil is a low rectangle of dry stone, easy to walk past. From the air it is a court hundreds of metres long. More than a thousand of them cross north-western Arabia, and the ones that have been dug held horns, not houses.

Why this is extraordinary

Across north-western Saudi Arabia, more than a thousand long rectangles of dry stone are known, from 20 to 620 metres on the long side, spread over about 200,000 square kilometres. People first recorded them in the 1970s. A systematic study came in 2017. The head is a platform, often 10 to 50 metres long and under about 1.2 metres high, with a chamber that may have a cap of stones or a blocked door less than half a metre wide. The long walls are only about 0.3 to 0.5 metres high. One court measured on the Harrat Khaybar lava is 525 metres long, built of basalt, and estimated at about 12,000 tonnes if a fifth of the volume is void.

The dated example sits 55 kilometres east of AlUla, 140 metres long, sandstone, head to the north-east under an overhang. Its chamber held horns and pieces of skull, mostly cattle, with sheep, goat and gazelle, set around an upright stone. A count gives 260 faunal remains and no mandibles, so the deposit was a choice of heads, not a dumped carcass. No human bone was in the chamber. A later cist cut the head and did hold a person. Radiocarbon puts the use from 5307–5002 calibrated BC to 5056–4755, the dates clustering near 5000 BC. That is earlier, by the published comparison, than the cattle monuments at Shi’b Kheshiya in Yemen, about 4400 BC.

A pen is hard to defend. Cattle step over a wall of half a metre, and nothing in these courts looks roofed or lived in. A solar calendar is also hard to defend. Solstice and equinox sunrises fall at known bearings. These rectangles do not share those bearings. They run across the slope, or along the long edge of a mesa, and on flat lava they do not agree with each other. The sky was available. The builders followed the ground.

What this changes: if the courts are ritual and the dates hold, monumental stone and a cattle rite in north-west Arabia belong to the late sixth millennium BC, before the kingdoms and before the tombs that later filled this desert. If they are mainly marks of range, the horns are how a claim was kept, and the empty length of the court is the monument.

Why it matters

If they are ritual courts of the late sixth millennium BC, monumental stone building in Arabia starts with cattle offerings, earlier than the southern cattle rites they have been set beside. If they are territorial marks that happen to contain offerings, the cult is real and the long walls are a claim on ground.

The form is two short platforms joined by two long walls, sometimes with a dividing wall. Lengths run from 20 to 620 metres. The head is a rubble platform, often 10 to 50 metres long and 0.3 to 1.2 metres high, with a chamber. Long walls are only about 0.3 to 0.5 metres high. Doorways can be under half a metre wide, and some were blocked. More than 1,000 are known across about 200,000 square kilometres, between about 23 and 28 degrees north and about 37 and 43 degrees east.

Photo tour

1 photograph, unedited. On a phone, swipe the row. Choose one to look closely.

No openly licensed ground photograph of the excavated chamber is in this room. The aerial is a different mustatil, published as a labelled figure.

What is still unexplained

These sit beside the consensus account, with the same visual weight. None of them is marked solved.

Open Horns around an upright stone A mustatil 140 metres long, 55 kilometres east of AlUla, was excavated with its chamber intact. Horns and skull parts lay in layers around a central upright, read as a betyl, a standing stone treated as a focus. Cattle dominate. Sheep, goat and gazelle are present. No human bone and no living floor were found in that chamber. A later count of the same monument gives 260 faunal remains, almost all from the skull, and no mandibles. Radiocarbon on the mustatil runs from 5307–5002 to 5056–4755 calibrated BC. A small cist cut into the head later and held human bone, so a grave came after the court, it was not the court’s purpose. Eleven dates from the site cluster near 5000 BC.

Open Too low to be a pen An older guess was that the rectangles held animals. The long walls are 0.3 to 0.5 metres high, the entrances are narrow, nothing was roofed, and no occupation debris is reported. A pen that a cow can step over is not a pen. The horns are selected body parts, not a herd that died in a corral. That pushes the pen reading down. It does not by itself say whether the rite was an offering, a feast, or a mark that this valley was spoken for.

Open No shared direction toward the sun A solar monument shares an azimuth. At this latitude a solstice sunrise is well north or south of east, and an equinox sunrise is due east. You would see the same bearing repeated. The survey found neither a preference for the cardinal points nor an aim at prominent landforms. On a hillside the long axis is perpendicular to the slope. On a narrow mesa it takes the longest edge. On the flat lava of Harrat Khaybar the directions look unsorted. One excavated example, the 140-metre court, lies roughly southwest to northeast, with the head to the north-east under a rock overhang, a local fit and not a rule for the class. A sky reading of these courts is not supported by a published common azimuth.

Location and dating

Location: The pin is 55 kilometres due east of AlUla town, the distance given for the excavated mustatil IDIHA-F-0011081. The paper used here does not print a GPS point, so the pin is that distance, not a surveyed mark. Other mustatils, including the one in the aerial, are the same class and are not this coordinate.
Period label: Current consensus — open to revision Late Neolithic, late sixth millennium BC into the early fifth. The excavated court’s modelled dates run from 5307–5002 to 5056–4755 calibrated BC.
How this dating was done: Radiocarbon on cattle horn and tooth, bioapatite and collagen, calibrated with IntCal20. The architecture is compared across the region. The orientation study is a survey of axes, not a sky chart.
What could change it: A date from the base of a wall, not from the horns placed in the chamber, and a published table of azimuths if anyone still wants to test a solstice.

Published ranges, drawn as bars. A bar is a claim about a span, not a point.

Excavated court near AlUla

5307 BCE5000 BCE4755 BCE

Excavated court near AlUla. Modelled span from 5307–5002 to 5056–4755 calibrated BC for the site, dates clustered near 5000 BC.

What is actually there

More than 1,000 mustatils, 20 to 620 metres long, across about 200,000 square kilometres.
Long walls about 0.3 to 0.5 metres high.
One ground-surveyed example on the Harrat Khaybar is 525 metres long and estimated at about 12,000 tonnes of basalt.
Cattle horns and skulls around a betyl. No human remains in the excavated chamber.
Axes follow slope and bedrock, not a shared sunrise.

Open questions

Question	Status
Is the long court a temple, a territorial mark, or both?	Open. The horns are ritual deposits. The repeated plan across a huge region is what a shared claim on land would also look like.
Were they ever aimed at the sun?	Open against it. The published survey finds no cardinal preference.

The current narrative

Where publication stands today. It is the best-supported reading, and it is open to revision. It is not a verdict.

The current narrative — open to revision

Ritual courts for a cattle cult

Mustatils are open ritual monuments of the late sixth millennium BC. People brought horns and skulls, mostly cattle, to a standing stone in the chamber. They are among the earliest monumental stone traditions in Arabia.

What supports this

The intact chamber east of AlUla held stratified horns and cranial pieces around a betyl, with radiocarbon in the late sixth millennium BC.
Walls too low to pen cattle, narrow doors, no roof and no domestic debris remove the corral reading.
The same plan repeats from AlUla to Khaybar, which a shared rite explains.

What pushes back

A repeated plan over 200,000 square kilometres is also what territorial markers look like, and the southern cattle sites have been read that way.
The betyl and the “offering” are interpretations of a deposit. The bones are certain. The theology is not.
Later cists with human bone show the monuments were reused. A reuse can be mistaken for the original purpose if the layers are mixed. At the dated court, the cist cuts the head, so it is later. That care has not been repeated at every site.

What would settle it

It is the centre because the horns, the betyl and the dates are from an intact chamber, and the pen reading fails on the height of the walls. It moves if a large sample of courts turns out to be empty of deposits and still built to the same plan, which would favour a mark over a rite.

Emerging narratives

Other published readings. Each may be right. None is adopted here unless the evidence carries it.

An emerging narrative

Marks of pastoral range that also received offerings

The long rectangle is a claim on grazing and on passage. The horns are a rite attached to that claim, as at later cattle sites in southern Arabia, not a temple type standing apart from land.

What supports this

The plan is consistent and widespread, which a shared territorial habit can produce.
Size is the point a mark needs. A 525-metre court of about 12,000 tonnes is visible as labour even when the walls are low.
Southern cattle monuments around 4400 BC have been tied to territory, feasting and control of herds. The northern courts would be an older cousin of that idea, by about 900 years on the published comparison.

What pushes back

The chambers hold selected skulls and horns, including wild gazelle with the cattle, which is a ritual menu more than a boundary stone.
Orientations follow the slope, not a border or a view of a landmark. The survey says there is no aim at prominent landforms.
No occupation and no roof make them poor herding stations. A mark can be empty. It does not explain the betyl.

What would settle it

It stays because the scale and the sameness ask for a social job beyond one family’s altar. It is not the centre while the only intact chamber is a deposit of horns.

Should the current view be reevaluated?

Open to revision

The date and the horns are strong. Whether the court is first a shrine or first a territory is not settled, and the sun is not in the plan. Open keeps the rite and the range together.

What drives this

More than 1,000 rectangles, up to 620 metres, walls a few tens of centimetres high.
Cattle horns around a betyl, dated near 5000 BC.
No shared cardinal azimuth.

This is the museum’s own weighing of the evidence. It is not a visitor vote, and it is separate from the slider.

Weigh it yourself

Anonymous. Nothing here is stored except on this browser. Your lean does not change the assessment beside it.

Ritual courts for a cattle cult Unsure Marks of pastoral range that also received offerings

If the contested reading is right

What would follow if a reading that is not the consensus turned out to be true, and what could settle it. This explains the stakes. It does not adopt the reading.

They are territorial marks with a cattle rite attached

The earliest monumental stone in north-west Arabia would be a pastoral claim, and the cult would be how the claim was renewed. A sky calendar would not be part of the building.

Why it is still worth a test

Dates on wall bases, not only on horn, and a count of how many undug chambers actually hold deposits.

Sources

Image credits

One mustatil from the air, structure IDIHA-F-0003301, with the paper’s labels. The head is the far platform, the long walls enclose a courtyard, and the base and entrance are at the near end. An I-type platform sits beside it. This is not the excavated monument whose bone was dated. . Kennedy and co-authors, CC BY-SA 4.0. Shown unedited.

← Saudi Arabia World map