Everything Iran remembers of Sasanian triumph is carved in stone, in the south-west, facing Rome: emperors kneeling at Naqsh-e Rostam, processions filling the gorge at Bishapur. This is the other front. On a green plain in the far north of Khorasan, a few minutes from the Turkmenistan border, a small columned hall of rammed earth carries the empire's war against the steppe — battle, hunt and fire — modelled not in rock but in gypsum plaster, with the names of the riders written beside them in Middle Persian. It lay under the ground for fifteen centuries. The stone in Fars weathered. The plaster held.
Dargaz (درگز) is about as far from the Iran of the postcards as the map allows: a low, warm, surprisingly green plain tucked behind the mountains of northern Khorasan, three and a half hours over the passes from Mashhad, with Turkmenistan beginning just beyond the fields. In antiquity this was not a backwater but a doorway — the plain of the medieval city of Abiward, on the frontier where settled Iran met the steppe. Doorways need guarding, and the Sasanian Empire, which faced Rome on one side of its world, spent much of the fifth century fighting for its life on this one against the Hephthalites — the confederations the Byzantine historians called the White Huns, whose horsemen came out of Central Asia, defeated Persian armies more than once — and once killed a Persian king.
In 1994, on the low mound of Bandian (بندیان) beside the Dorungar river, two kilometres outside Dargaz town, Iranian archaeologists led by Mehdi Rahbar began digging, and kept digging for six seasons. What came out of the ground was, by any measure, one of the most important Sasanian discoveries of the modern era — and one of the least famous. A hall of chineh, rammed earth, its roof once held up by four plaster-coated columns; around its walls, in low relief, a programme of narrative scenes modelled in gypsum stucco: a cavalry battle, a hunt, dignitaries seated and standing, figures attending a fire altar. Beside the figures, pressed into the same plaster, run inscriptions in Middle Persian — naming, as the excavators read them, the men shown. Off the hall: a fire room with its plaster altar still in place, carved one-piece stone ossuaries for the bones of the dead, and a round funerary tower.
The material is the point. Sasanian figural art has come down to us almost entirely as rock relief — some thirty monumental carvings, nearly all in the south-west, nearly all facing the Roman world. Stucco was the empire's other voice: gypsum, mixed fast, modelled fast, painted, and fragile — architecture's makeup rather than its bones. Almost everywhere else it survives as fragments: a border here, a royal bust there. At Bandian a whole narrative programme stayed on its walls — scene after scene around a room, with captions — something the standard reference literature describes as being of dimensions previously unknown in Sasanian art. And it survives precisely because of what it is: when the building's roof came down, the earth that buried the hall sealed the soft plaster away from fifteen centuries of weather that has been steadily erasing the hard stone of Fars.
Who raised the hall, and for which victory, is still argued — the scholarship's attributions run from Yazdgerd I through Bahram V to Peroz and Kavad I, which is to say: the fifth century, the exact decades when this frontier was the most dangerous place in the Iranian world. The building's function is argued too, and honestly: the excavator published it as a fire temple, others have read an elite residence, and the newest excavations on the neighbouring mound describe a palace. What no one disputes is what you can see: a room where an empire under existential pressure recorded, in the cheapest and quickest of its arts, the enemies it feared most — and where that record outlived the empire, the enemy, and the argument.
The reliefs run around the hall in low relief at standing height — an art meant to be walked past, scene by scene, the way you would read. The upper registers are lost with the upper walls; what remains is the world at eye level:
Armoured horsemen ride down their enemies — the frieze read as a royal victory over the Hephthalites, the steppe power that dominated this frontier's fifth century and twice humiliated the empire. Here, for once, the empire is winning, and wanted it on record.
Riders at full stretch after game — the scene every Sasanian king wanted himself shown in, the aristocratic proof of nerve and horsemanship. The same subject fills the silver plates in the world's museums; here it is done in plaster on a frontier wall.
Standing and seated dignitaries, and figures flanking a fire altar — the scene that anchors the hall's religious reading. War on one wall, the fire on another: the two things the Sasanian state believed held the world together.
Beside the figures run inscriptions in Middle Persian, pressed into the stucco — read as the names and ranks of the men shown. Sasanian rock reliefs rarely tell you who anyone is. This room does. It is history with the labels still attached.
Off the columned hall, a smaller room holds a plaster fire altar, still standing where it stood — the strongest single piece of the excavator's case that this complex was a working Zoroastrian sanctuary, not merely a decorated hall.
Along the walls of a burial room lay carved one-piece stone ossuaries — receptacles for the cleaned bones of the dead, in proper Zoroastrian practice — and nearby, a round funerary tower, only partly excavated. The garrison's living and its dead kept close company.
Stand at Naqsh-e Rostam and you are looking at the empire's western face: a Roman emperor kneels before Shapur's horse, cut into a cliff at colossal scale, in stone meant to outlast everything — the Sasanian state talking to Rome, and to eternity, in its most expensive voice. The gorge at Tang-e Chogan is the same speech continued: processions, captives, triumph, all in rock.
Bandian is the same empire on its opposite face, and everything is inverted. The enemy is not a rival civilisation but the steppe — mobile, unglamorous, and in the fifth century far more dangerous: Hephthalite armies defeated Peroz twice and killed him the second time. The setting is not a royal gorge in the heartland but a garrison plain at the edge of the world. And the material is not cliff-stone but gypsum plaster — quick, cheap, local, fragile — the medium of a frontier that needed its victory recorded now, not in a generation of stonemasons' time.
Then the two fates crossed. The stone of Fars has stood in the open for seventeen centuries, and the weather has been slowly eating it the whole time. The plaster of Bandian was buried by its own collapsing roof within a few generations — and the burial preserved it. The empire's loudest voice is fading in the open air. Its quietest one waited under a field, and came back word for word.
Bandian asks nothing of your body and a fair amount of your itinerary: the visit itself is a stroll under a roof, but the roof stands on a remote plain three and a half mountain hours from Mashhad, in a border district most travellers never think of. Its weight is all on the Legacy scale, and concentrated in one criterion — there is simply nothing else like this room in Iran.
Every Sasanian monument you have ever seen taught you to look up. The tombs at Naqsh-e Rostam hang a cliff-face over your head; Shapur's triumphs are cut at twice life size; the fire temples crown hills. Bandian retrains you in one step through the door. You are under a modern roof, in the soft light of a protected dig, and the empire is at your elbow — horsemen the height of your forearm riding along a wall of pale plaster, close enough to see the modelling of a bridle, the set of a shoulder, the tool-strokes of a craftsman who worked fast because the frontier did not wait.
And then the thing that no rock relief in Iran will give you: writing, right there beside the figures. You cannot read it — almost nobody can — but you can see instantly what it is: names. Someone walked a visitor around this room fifteen centuries ago the way a host walks a guest past photographs in a hallway: this is the commander, this is the battle, this is the fire we keep. Every other Sasanian monument performs at you. This one introduces you.
Walk back out onto the plain. It is green, and flat, and quiet, and a few fields away is a border fence with a different country behind it — the direction the horsemen in the plaster are riding, the direction the fear came from. The stone monuments of Fars were built to be seen forever and are slowly failing. This little hall of mud and gypsum was seen by a garrison for a few generations, fell, was forgotten — and it is the one that came back whole. You are standing in the empire's quietest room, and it turned out to have the longest memory.
An empire that answered Rome in cliff-stone answered the steppe in gypsum — and the fragile answer is the one that survived.
The Dargaz plain is one of the greenest corners of Khorasan in spring — young wheat to the horizon, the mountains still carrying snow, and the drive over the passes from Quchan at its finest. The site is the same in any weather; the journey is not, and this is the journey's best version.
Clear, calm and empty. Harvest colours on the plain, comfortable temperatures, and the best chance of having the hall entirely to yourself — which, in truth, you will have most days of the year anyway.
Dargaz sits low and runs genuinely hot in high summer. The saving grace is that the visit itself happens under a roof; come early, see the hall in the cool hours, and treat the drive as the hot part of the day.
Winters are cold, with occasional snow over the pass and a mild plain below — the emptiest season at the site. The shelter is unheated; bring a proper jacket, and you will have the fifth century entirely to yourself.
The wonder of this place is above. What follows is the planning detail — gear, logistics, and the questions people ask — tucked away so you can open only what you need.
Bandian is far from everything except Dargaz — which is precisely its historical point. Prices move with the rial, so treat any figure as an order of magnitude.
On the Dargaz plain in the far north of Razavi Khorasan, about two kilometres north-west of Dargaz town, close to the Turkmenistan border. Dargaz is roughly 220 km north of Mashhad — about three and a half hours by road over the mountains via Quchan. Buses and shared taxis connect Mashhad and Dargaz; from the town, the site is a five-minute taxi ride.
A pisé-built columned hall of the fifth century CE whose walls carry narrative stucco reliefs — a battle, a hunt, figures at a fire altar — with Middle Persian inscriptions naming the people shown; a fire room with its plaster altar still in place; carved one-piece stone ossuaries; a round funerary tower; and, on the neighbouring mound, the remains of a large building read by its newest excavators as a palace. The excavated area is roofed and kept as a site-museum.
Sasanian figural art survives almost entirely as rock reliefs in the south-west of Iran. Bandian is different in every way: it is plaster, not stone; it is narrative, running scene after scene around a room; and it is captioned — the names of the figures are written in Middle Persian beside them. The Encyclopædia Iranica calls it a programme “of hitherto unknown dimensions” in Sasanian art.
The steppe. The reliefs are read as commemorating Sasanian victories in the fifth-century wars against the Hephthalites — the power the Byzantine historians called the White Huns — whose raids out of Central Asia made this frontier the empire's most dangerous. Which king's victory is shown is still argued: attributions across the scholarship range from Yazdgerd I to Bahram V, Peroz and Kavad I.
Both readings are on the table, and the argument is part of the site's interest. The excavator, Mehdi Rahbar, published the columned hall as a fire temple, and a fire room with its altar supports him. Others have read an elite residence, and the newest excavations on the adjacent mound describe a palace — with one space tentatively interpreted as a Mithraeum. What nobody disputes is the date, the material and the frontier.
Yes — the excavated hall is protected under a roof and run as a working site-museum, with walkways around the dig, so unlike most excavations you actually see the stucco in place. Allow 45 minutes to an hour and a half. Many visitors pair it with Tandoureh National Park in the mountains above the plain — a separate outing with its own inexpensive permit and ticket.
Spring, without much competition: the Dargaz plain is one of the greenest corners of Khorasan from March to May, and the mountain road from Quchan is at its finest. Autumn is the quiet second-best. Summer on this low plain is genuinely hot, and in winter the pass can occasionally carry snow.
Bandian opens a corridor this collection has been waiting to draw: deep Khorasan, the frontier province where Iran has always met Central Asia. Its natural companion is Ribat-e Sharaf, the Seljuk caravanserai stranded in the desert on the Sarakhs road — the same frontier, seven centuries later, when the fear had given way to commerce and the garrison hall's successors were palaces for merchants. South, at the province's other end, the Qasabeh qanat of Gonabad is Khorasan's third act of quiet genius: a 300-metre-deep river dug by hand, still running. And for the other half of Bandian's own argument — the empire's stone voice — read it against Naqsh-e Rostam and Tang-e Chogan, where the same dynasty addressed Rome at colossal scale, and against Takht-e Soleyman, where its greatest fire burned beside a bottomless lake.
The finest Silk Road monument in Iran, alone in the desert between Merv and Nishapur — dazzling Seljuk brickwork built for merchants on the frontier Bandian's garrison once watched. Read the article →
The mountain wall directly above the Dargaz plain: juniper slopes, wild ibex and urial, and one of Iran's strongholds of the Persian leopard. Its permit and ticket are inexpensive — a separate outing from the plaster hall below.
The medieval city whose plain this was — a name that echoes through Khorasani history and poetry, its low ruins scattered toward the border. Little is excavated; the setting, not the stones, is the visit.
Fifteen kilometres east of Serakhs, in southern Turkmenistan: a fifth-to-seventh-century fire temple whose stucco is compared stylistically to Bandian's. The same school, the same frontier — today on the far side of a border neither temple can see.
Come in late April, when the plain is at its greenest and the passes are clear. See the hall in the morning light, then drive up into Tandoureh in the afternoon and look back down: the fields, the town, the low mound with its modern roof, and beyond them the fence running east to west. Somewhere under that roof, horsemen the colour of ivory are riding toward the border at full gallop, exactly as they have been for fifteen hundred years — the empire's quietest victory, still in progress.
Untamed Iran prefers official, scholarly and first-hand sources, and is careful to separate what is established from what is still argued. Bandian is unusual among lesser-known Iranian sites in having a genuine scholarly literature — excavation reports, epigraphic studies and reference-work treatment — alongside the usual travel summaries. The following are the sources this page rests on:
Facts last reviewed July 2026. Established: a Sasanian complex of the fifth century CE on the Bandian mound, about 2 km north-west of Dargaz in Razavi Khorasan, by the Dorungar river, near medieval Abiward; excavated from 1994 under Mehdi Rahbar across six seasons; a rammed-earth columned hall with narrative gypsum stucco — battle, hunt, figures at a fire altar — and Middle Persian inscriptions beside the figures; a fire room with altar, stone ossuaries and a round funerary tower; the excavated hall roofed and run as a site-museum. Read differently by different sources: the royal attribution — Yazdgerd I, Bahram V, Peroz and Kavad I all appear across the literature; this page holds to “the fifth century”. The building's function — fire temple (the excavator's published reading, supported by the altar), elite residence (other scholars), and, for the adjacent mound, a palace with a tentatively identified Mithraeum (the newest excavation report). Details of the ossuaries (their number and decoration) rest on heritage and travel summaries rather than the primary reports available to us. Figures: the coordinates and the 470 m elevation follow the site-museum's published data; the ~220 km / three-and-a-half-hour Mashhad–Dargaz figures are road estimates. Deliberately not claimed: that Bandian is unique on Earth — its standing rests on the published scholarly assessment quoted above, not on a superlative of ours; and precise opening hours.