the Murghab delta, north of Mary, Turkmenistan

Gonur Tepe

About 2400 to 1600 BC (the Oxus civilization’s main city, with a ritual drink still argued)

Mud-brick walls of a ruined complex, with two pale pottery vessels set on the excavated brick. — Mud-brick walls at Gonur Tepe. Two ceramic vessels sit on the excavated brick. The photograph does not show whether they were found in those spots or set there so the plan can be read. The scale is a city of brick, not a single room. David Stanley, CC BY 2.0. Shown unedited.

About 60 kilometres north of Mary, in the old delta of the Murghab, a mud-brick city covers some 55 hectares. It was the main settlement of the Oxus civilization, also called the Bactria–Margiana complex, between about 2400 and 1600 BC. A palace, water pipes, and a grave with a foal on a wagon are the surprises. A ritual drink is the argument.

Why this is extraordinary

North of Mary, in the dry channels of the Murghab, a grey hill turned out to be a city. Pottery came off the surface in 1972. Digging from 1974 onward opened a mud-brick capital of the Oxus civilization, the culture also called Bactria–Margiana, sitting between Mesopotamia and the Indus in the centuries about 2400 to 1600 BC. The whole ruin is about 55 hectares. Gonur North, about 330 by 460 metres, holds a palace of about 150 by 140, two courtyards, an axial corridor, and ceramic pipes at the door. Gonur South, about 130 by 120 metres, is the later enclosure, about 1900 to 1600 BC, with round towers. West of the north city, a necropolis of more than 3,000 graves took adults and older children. The very young were buried by the houses. A restudy counts five building phases, from a foundation as early as about 2500 or 2300 BC to an abandonment by about 1500, with a great fire in the palace around 1900 BC in the middle of that story.

The outside world is in the graves. A nearly complete foal lay on a four-wheeled wagon with bronze-rimmed wheels in grave 3200, radiocarbon about 2250 BC, which has been set roughly a century and a half before the wagon burials of the Sintashta steppe. Ivory of the Asian elephant belongs to the last royal phase and is compared with Mohenjo-daro. Mosaics on chests are compared with Mari and with Ur. Steppe pottery shows up as the city ages. The river, meanwhile, moved west, and that is the plainest reason the brick was left.

The argument that will not sit quietly is the drink. Vessels were reported with traces of cannabis, poppy, and ephedra, and a mortar and pestle were read as the press for the soma of the Rigveda and the haoma of the Avesta. Fire altars were tied to rites that later texts give to Zoroastrian practice, well before the century usually given to that religion. The catch is the gap. A plant in a pot can be medicine, food, or a rite, and the chemistry does not say which. The poems are centuries later, and Gonur has not produced a readable text that names the drink. No solstice or star azimuth for the palace corridor, or for the cruciform building in the south enclosure, was found in the sources used here. An axis can be ceremony without being a sightline. What this changes: the map of Bronze Age states gains a capital in the Murghab, with Indus ivory and a wagon. Whether that capital was already practising the rite the later poems describe is a separate claim, and it still hangs on residues and on names the brick does not spell.

Why it matters

The Oxus cities sit between Mesopotamia and the Indus and were missing from the old map of Bronze Age states. If Gonur’s altars and plant traces are the ancestor of the later haoma and soma rites, a religion known from poems is already in the brick. If they are a local practice with local plants, the city is still a capital, and the poems are not required.

The local name means grey hill. Pottery was picked off the surface in 1972, and excavation ran from 1974 for decades. The plan has three parts. Gonur North is an oval enclosure about 330 by 460 metres, in use from about 2400 to 1900 BC, with a monumental palace of about 150 by 140 metres, courtyards, an axial corridor, storerooms, and rooms read as audience halls. Water was piped in ceramic near the entry, which has been read as washing before you went in. A royal graveyard sits in the southeast of that enclosure. Gonur South, about 130 by 120 metres, was lived in from about 1900 to 1600 BC, inside two rings of wall with round towers. A large necropolis of more than 3,000 graves lies about 200 metres west of the north city, some 10 hectares, and it took adults and youths over about eight. Children were buried by the houses. A stratigraphic restudy published the site in five stretches: foundation about 2500 or 2300 to 2000 BC, the palace until a great fire about 2000 to 1900, a restored palace to about 1700, common use of the palace to about 1600, and abandonment by about 1500. Over 150 other settlements are known in the same delta.

Photo tour

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

No openly licensed plan of the palace axis was found, and no sky azimuth is claimed. The one photograph is the mud-brick ruin with vessels set on the walls.

What is still unexplained

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

Observed A foal, a wagon, and ivory In grave 3200 of the royal necropolis, a nearly complete foal lay on a four-wheeled wooden wagon with bronze bands on the wheels. A radiocarbon result for that burial is about 2250 BC. A comment on the find places it about a century and a half before the famous wagon graves of the Sintashta culture on the steppe. Another grave, 3210, held a stone figure read as a horse with a saddle. Grave 3310 held parts of a stallion, head and hindquarters missing, read as a cult burial. Ivory of Asian elephant turns up in the last phase of the royal necropolis, in levels set beside late Altyn Depe and beside Mohenjo-daro. Mosaics on wooden chests have been compared with work at Mari in Syria and with the Royal Cemetery at Ur. So the city is not a backwater. It is tied west, south, and, on the horse, north. The catch on the horse is the one bone. A foal on a wagon is a domestic animal if the identification holds, and one grave is a thin basis for saying the steppe learned the wagon from Margiana, or the other way around. Steppe pottery does occur in the later layers. How much of it, and whether it means a takeover or a trade, is still counted differently from survey to survey.

Open Plants, fire, and a drink from later poems The excavator of the palace reported traces of cannabis, poppy, and ephedra in vessels, and a press of mortar, pestle, and a stone with a hemispherical knob, and read them as the ritual drink called soma in the Rigveda and haoma in the Avesta. Fire altars in the temples were tied to rites that later belong to Zoroastrian practice, centuries before the usual dates given for that religion. In Gonur South a walled temenos and an unfinished cruciform building with twelve round towers were given the same cult reading. The catch is the distance. A residue in a pot is a plant. A plant is a medicine, a food, a drug, or a rite, and those are not separated by chemistry alone. Soma and haoma are names in poems set down much later. Using the poems as the label is a hypothesis about continuity. Refusing the poems is a hypothesis that a local rite needs no later text. Neither side has a written recipe from Gonur. No long inscription has been read from the palace. And no solstice or star azimuth for the palace axis, or for the cruciform building, turned up in the sources used here. An axial corridor and water at the door can be ceremony without being a sightline. The abandonment is plainer. The Murghab shifted west, and the city ended.

Location and dating

Location: In the old delta of the Murghab, about 60 kilometres north of Mary, ancient Merv. The coordinate is the ruins. Gonur North and Gonur South are parts of one site, not two cities.
Period label: Current consensus — open to revision Bronze Age, about 2400 to 1600 BC for the main life of the city, with a foundation that a 2019 stratigraphy starts as early as about 2500 or 2300 BC and an abandonment by about 1500. This is the Bactria–Margiana Archaeological Complex, the Oxus civilization.
How this dating was done: Radiocarbon on burials, including the wagon grave near 2250 BC, and the pottery and building sequence of the five published phases. The ritual-drink reading is a chemical trace plus a later text. It is not a date.
What could change it: A published identification of the plant remains that separates ephedra, cannabis, and poppy from look-alike residues, with the vessel’s place in the palace drawn. A sunrise azimuth for the palace axis, if one is ever measured, would be new. None is in the sources used here.

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

Main city

Five phases

Wagon grave

2500 BCE2000 BCE1500 BCE

Main city. The span usually given for Gonur. North city about 2400 to 1900 BC, south city about 1900 to 1600.
Five phases. The 2019 internal chronology, foundation to abandonment. The early end is written as about 2500 or 2300 BC.
Wagon grave. About 2250 BC for the foal and the wagon in grave 3200. A single burial, not the foundation date of the city.

What is actually there

The ruins cover about 55 hectares, with a palace about 150 by 140 metres.
Gonur North is about 330 by 460 metres. Gonur South is about 130 by 120.
The western necropolis has more than 3,000 graves and excluded young children.
Grave 3200 held a foal on a four-wheeled wagon, radiocarbon about 2250 BC.
Cannabis, poppy, and ephedra have been reported in vessels and read as soma or haoma.
No solstice or star azimuth for the palace was found in the sources used here.

Open questions

Question	Status
Are the plant traces the drink named in the Rigveda and the Avesta?	Open. The plants are a report from the vessels. The poems are much later. A local rite does not need those names, and those names do not need to be false.
Did the horse and wagon here come before the steppe wagon graves?	Open. One foal at about 2250 BC has been set a century and a half before Sintashta. One grave is a thin sequence.

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

A Bronze Age capital between the rivers

Gonur is the principal city of the Oxus civilization, a mud-brick state in the Murghab delta from about 2400 to 1600 BC, trading ivory, metal, and style with the Indus, Iran, and Mesopotamia, and ending when the river moved.

What supports this

The scale is a city. Fifty-five hectares, a palace, twin enclosures, and more than 3,000 graves are not a camp.
Ivory in the late royal graves is set beside Mohenjo-daro, and mosaics are set beside Mari and Ur. The contacts are objects, not a guess.
A restudy gives five building phases from a foundation in the later third millennium to abandonment by about 1500 BC, and the river’s shift west is the plain reason the ground was left.

What pushes back

“Capital” is a word. Gonur is the largest dug city of the complex. Other delta towns exist, and a political capital is more than the biggest ruin.
The horse and the steppe pottery show the city was not sealed off. A takeover from the north is still argued for the end.
Without a readable script, the name the inhabitants used for themselves is not on the brick. Margush is a later scholarly name.

What would settle it

It is the centre because the plan, the graves, and the outside objects are excavated. The ritual names and the horse’s priority are the parts that can move without shrinking the city.

Emerging narratives

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

An emerging narrative

Fire altars and the drink of the later poems

The temples, the press stones, and the traces of ephedra, cannabis, and poppy are the rite that Indian and Iranian poems later call soma and haoma. The fire altars are an ancestor of Zoroastrian practice, earlier than Zoroaster’s usual century.

What supports this

Vessels with those three plants, and a mortar-and-pestle set, were reported from the ritual rooms and compared with a pressing kit.
Water at the palace door fits a washing before entry, which is the sort of act a controlled rite uses.
The same excavator found the architecture repeating at other Margiana sites, so it is not one room’s accident.

What pushes back

Plant traces are not a recipe. Food, medicine, and a rite can share a pot.
Soma and haoma are texts from centuries later. Continuity is a long bridge, and a local cult can burn fires without being that bridge.
No inscription from Gonur names the drink or the fire. The label is imported from the poems.

What would settle it

It stays because the plants and the altars were published from the rooms, not invented from the poems alone. It is not the centre while the poems do the naming and the chemistry does not choose a rite over a kitchen.

Should the current view be reevaluated?

Open to revision

The city, the palace, the graves, and the outside connections are firm. The soma reading and the priority of the wagon are real published claims with thin locks, one set of residues and one foal. Open keeps the capital and leaves the drink and the horse visible.

What drives this

A 55-hectare mud-brick city, about 2400 to 1600 BC, with a palace and more than 3,000 graves.
A foal on a wagon at about 2250 BC, and ivory tied to the Indus orbit.
Reported ephedra, cannabis, and poppy, read as a later ritual drink, with no text from the site.

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.

A Bronze Age capital between the rivers Unsure Fire altars and the drink of the later poems

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.

The vessels hold the drink later called soma and haoma

A rite known from the Rigveda and the Avesta would have a Bronze Age kitchen in Margiana, and the fire altars would be taught as older than the usual date for Zoroastrian practice. The city would be a religious capital as well as a trading one.

Why it is still worth a test

Residue identifications published with the vessel, the room, and the phase, set beside the pressing stones, without borrowing the poems’ names until the plants match a recipe.

Sources

Gonur Depe Wikipedia, drawing on the excavation reports for plan and phases
Archaeology and language, the Oxus civilization The published discussion of BMAC chronology, Andronovo contact, and the ritual reading

Image credits

Mud-brick walls at Gonur Tepe. Two ceramic vessels sit on the excavated brick. The photograph does not show whether they were found in those spots or set there so the plan can be read. The scale is a city of brick, not a single room. . David Stanley, CC BY 2.0. Shown unedited.

← Turkmenistan World map