108–110, 271, 2024: the colony, the fortress, the heritage # Two Sarmizegetusas The [Sarmizegetusa](https://mariuscomper.uk/sarmizegetusa-mituri/) in the mountains is Dacian: Decebalus's fortress, with stone sanctuaries. But 40 km west of it, in the plain of Țara Hațegului, the Romans walled a new city after the conquest — and gave it the same name, Sarmizegetusa. Two Sarmizegetusas. One in the mountains, one in the plain. The one in the plain is called Rome: Colonia Ulpia Traiana Augusta Dacica Sarmizegetusa, the symbolic, fiscal-administrative and religious capital of Roman Dacia, with a forum, an amphitheatre and some 20,000 inhabitants. We walk it together, from the founder's plough to the gates blocked with tombstones: on the left you see the year and what is being walled, played or defended. And at the end we do what we did at Argeș, at [Hunedoara](https://mariuscomper.uk/castelul-corvin/), at Voroneț, at Peleș, at Constanța and at [Adamclisi](https://mariuscomper.uk/adamclisi/): we grade the legends — what the stone says, what the paper says, what the museum says and what no record says. ## Where Ulpia stands In Hunedoara county — like [Corvin](https://mariuscomper.uk/castelul-corvin/) — in the west of Țara Hațegului, on the upper Breazova, some 8 km from the Poarta de Fier a Transilvaniei pass, the way through to the Banat. The ruins stretch some 1500 by 1000 metres, under today's village and east of it; the west wall lies under the village, the north one under an unexcavated earth bank. On record: RAN 91063.01, LMI HD-I-s-A-03205, class A, 2nd–3rd centuries. The map below takes you from the country to the forum. The amphitheatre The forum The museum The amphitheatre The forum The museum Outlines: Natural Earth (public domain). Coordinates: OpenStreetMap / site plan. The founding · The plough and the brackets ## Who laid the foundation stone? Not Trajan. The emperor ordered from afar; the stone was laid by governor Decimus Terentius Scaurianus, in his name — so says the forum's foundation inscription, found broken, with missing parts that rob us of an exact dating. That is why the years come as a span: 108–110, say the guides; 109–110 even, says the Deva museum, matching Scaurianus's term in office. The exact year stays unknown — and this page keeps it so. The ritual survives on a bronze sestertius, a foundation coin type: Trajan's head on the obverse, and on the reverse — probably — the governor with his head veiled in the toga, driving a white ox and a white cow to trace the sacred furrow, the pomerium, along which the walls would rise. The same ritual legend credits to Romulus at Rome. And the colony's full name reads with square brackets — the archaeologists' mark that the missing letters are their restoration: condita colonia [Ulpia Traiana Augusta] Dacica [Sarmizegetusa] per [D. Terenti]um Scaurianum [legatum] eius pro pr. Letters for Trajan, at the village museum: blocks from the outer arch of the tetrapylon at the forum entrance. Photo: Carole Raddato, via Wikimedia Commons. The forum · Where the roads crossed ## What was judged at Sarmizegetusa? Everything. Here the two main streets crossed, and the crossing point — the locus gromae — was marked with a marble base and a small altar. You entered the forum court through a monumental four-pillared gate, a tetrapylon, a double triumphal arch, on whose pediment the city's foundation inscription must have stood. The court was clad all over in marble, with socles for statues of emperors in gilt bronze or marble; around it, porticoes with marble colonnades some 4 metres tall, under tiled roofs. From the court you passed into the basilica, the building that towered over all: east and west, a podium each — the tribunalia — where citizens held their speeches and where the two mayors, the duumviri, judged. Under the east tribunal stood a carcera — not quite a prison, more a remand cell. Next to it, the curia, where the decurions debated the city's business, and two vaulted rooms, the aeraria, the city's treasury. Today's forum is the result of several phases — timber, Trajanic stone, macellum, Hadrianic completion — and what Daicoviciu and Floca took in 1924 for a “Palace of the Augustales” later proved to be the very forum of Colonia Dacica. The forum today: walls, columns and socles under the Țarcu mountains. Photo: Razvan Socol, via Wikimedia Commons. The walls · What came before ## What was the city walled on? On a wall rectangle some 600 by 500 metres, about 5 metres high, with round corner towers — two uncovered, south-east and north-east — enclosing some 33 hectares, plus another 60–80 hectares of building outside the walls: baths, suburban villas, the amphitheatre, brick kilns, necropoleis. CIMEC says a camp of Legio V Macedonica lay beneath; newer research raises an eyebrow — stamped legionary bricks may come from builders, not from a camp — so this page keeps the legion as a debated hypothesis, not a certainty. The same goes for the first people: the city was a deducta colony, and CIMEC speaks of a deductio veteranorum — veterans settled after the wars. But a new study counts barely two veterans securely dated to the first wave. Veterans were among the colonists, but they were not the only colonists: next to them came people from all over the empire, including Palmyra, gods and all. The name · Face to face ## Who called it Sarmizegetusa? Two papers, two answers — this page sets them face to face. The old version, from guides and encyclopaedias: under Hadrian (117–138) the city receives its full name, Colonia Ulpia Traiana Augusta Dacica Sarmizegetusa, with the metropolis epithet added later, in 222–235. The new version, from modern epigraphic reconstruction after the 1990s fragments: the full name may have stood there from the very beginning, from the founding. Today's historians read the gesture as a taking of possession: the Dacian capital had been wiped off the map, and its name was moved 40 km west, onto the signboard of the new Roman capital. Mind it: the motive is their reading, not an ancient record. Not even the name's meaning is settled: what Sarmizegetusa means is not exactly known, and the split into zermi (rock, height) and zeget (palisade, fortress) — “the fortress on the rock”, “the high fortress” — is a mere hypothesis, credited to I. I. Russu. The exact meaning stays unknown — and it proves nothing about the Roman city standing on any high rock: it stands in the plain. The arena · 5,000 spectators ## How did you get into the games? Some 100 metres north of the walls, the 88 by 69 metre ellipse: the amphitheatre, with its 66 by 47 metre arena and stands for some 5,000 spectators. Seats were taken in very strict order: in the front rows, on stone benches — some with reserved names on the backrests, still legible — the city's aristocracy, the ordo decurionum; then the knights, the businessmen; in the last rows, the plebs and the women. On sand or sawdust they played gladiatorial fights, drama, mime, sport — and after the games the surface was washed with water, drained through a great channel cutting the arena from west to east. In the channel, midway, a strange chamber: it housed, probably, a machine — the pegma — which the museum says an inscription attests. What it looked like, nobody knows; picture a wooden platform, like a lift, raising into the arena a cage of beasts or gladiators. And at the entrance Nemesis waited: her temple stands right at the amphitheatre gate, and the inscriptions found inside show whom it served — the goddess of luck and vengeance, whom gladiators needed so badly. Of particular fights, with beasts and executions and names, this page tells nothing: the paper does not give them. The 88 by 69 metre ellipse, with the arena channel in view. Photo: Popescu Camelia, via Wikimedia Commons. The fortress · The dead defend the living ## How did the arena become a fortress? Between 271 and 275, the Roman army and administration withdraw south of the Danube — the ancient sources speak of abandoning Dacia, not of evacuating Ulpia street by street, so this page tells no “departure day”. The settlement kept living on afterwards, into the 4th century. And sometime in the 4th century, the amphitheatre seems to have played another part too: by its gates were found tomb-lid fragments and a funerary stele, brought from the city's cemeteries to block the gates — the dead defending the living — and turn the arena into a fortress far easier to defend than the city walls. A hoard of small bronze coins, found in the stands, dates the blocking: the 4th century, when the city no longer had people enough to defend the walls, but could still put up resistance. The last limit: a late-4th to early-5th century building in the north-east corner — that far was the city inhabited in antiquity. Then the Slavic place-names — Teregova, Voislova, Bucova, Grădiște — and, in 1315, the village recorded as Britonia. It would be called Sarmizegetusa again only in the contemporary era. The arena channel today: the games' water drained here — and midway stood the machine. Photo: Grigroibu, via Wikimedia Commons. The gods · A dictionary, with traps ## In how many tongues did the city pray? On record, in one: Latin is the language of the inscriptions and of public life — from the foundation text to the names on the bench backrests. Greek shows up in religious inscriptions, around a debated eastern cult, Theos Hypsistos. And Aramaic? A trap: at Ulpia itself no Aramaic inscription was ever found — Palmyrene people yes, temple and gods included (Malagbel, Bel Hamon, Fenebal, Manavat), but written evidence in Aramaic at Ulpia, none. Whoever sells you the trilingual city sells you legend. Otherwise the pantheon is packed: Jupiter and Hygieia in votive statues, the Thracian Rider, Diana, the Danubian Riders, Mars, Minerva, Mithras with his temple in the south-west corner, the Syrian gods, Malagbel, Silvanus of the forests with his Silvanae, Liber Pater — and Aesculapius with Hygieia in the sacred area by the amphitheatre, the “listening gods”, shown as ears, to whom sufferers whispered their troubles; their temple may have doubled as a hospital, with first aid at the arena gate. And the building by the amphitheatre, called the “Schola gladiatorum”, stays so-called: no inscription proves its purpose. Nemesis's temple, at the amphitheatre gate: luck, justice, vengeance. Photo: Andrei Stroe, via Wikimedia Commons. The stones · Who took from the ruin ## Where did the walls go? Into churches. The Roman ruins served for centuries as a quarry of ready-cut stone, and the Deva museum lists the Țara Hațegului churches reusing Roman pieces: Densuș, Ostrov, Peșteana, Hățăgel, Tuștea, Sântămăria Orlea. Densuș reuses Roman pieces — that far does the paper go. That the church was walled specifically from the stones of the Ulpia amphitheatre is talk without a stone-by-stone study: for Ulpia the paper gives a single piece, one capital “said to” have reached Densuș. The rest is legend with a steeple. The stones also reached museums: Deva, Cluj, Bucharest, Lugoj — even Budapest and Vienna. And the other way round, the village museum, opened on 9 August 1982, keeps together what is left: coins, funerary stelae and altars, inscriptions, the votive statues of Jupiter and Hygieia, the riders, Diana, Trajan's bronze head, mosaics, glassware. In the courtyard, the lapidarium: the big marble, under the sky. Densuș: the stone church reusing Roman pieces — but we don't know from which walls. Photo: Mihai Raducanu, via Wikimedia Commons. Today · Who digs, who guards ## Whose is the city today? The museum's and the village's, at once. The ruins are dug by archaeologists — the thread starts with priest Mezerzius, gathering inscriptions from around 1495, runs through Zamosius and Marsigli, through the mosaic Ackner uncovered on 10 September 1832, through Mommsen — who gathered inscriptions now lost — through Király and Téglás, through Daicoviciu from 1924, through Floca in 1936 and again in 1948 at Hobița, through the 1973 resumption, and reaches the German team digging today in the sacred area, where it has already found a new temple. Since 2024 the section is a department of the Deva museum; the EU-restored amphitheatre has reopened its gates, and the project took in 2025 the prize of the National Restoration Triennial. And UNESCO? Here the page corrects its older siblings: both Sarmizegetusas are UNESCO. Regia since 1999, in the Dacian Fortresses of the Orăștie Mountains — and since 2024, both of them, as parts of the “Frontiers of the Roman Empire – Dacia” file, inscribed on 27 July 2024: Ulpia is component 1718-017, Regia is 1718-272. Two Sarmizegetusas, two listings for one, one for the other — and no “Regia yes, Ulpia no” holds any more. The arena today, with the museum behind: the yellow building has sheltered the inscriptions since 1982. Photo: Uhserban, via Wikimedia Commons. The legends · Each with its grade ## What is colony and what is talk? The records give the years: founding 108–110, no exact year, amphitheatre blocked in the 4th century, the village Britonia in 1315, the museum in 1982, UNESCO in 2024. The paper gives the people: Scaurianus, not Trajan; veterans among the colonists, not alone; Palmyrenes, gods and all; Mezerzius, Ackner, Mommsen, Daicoviciu, Floca, the German team. Legend says the plough was Trajan's, Densuș is from the amphitheatre and the city was trilingual — and the grade table sits below, at [myths](#mituri). The grading thread is held by [Hunedoara](https://mariuscomper.uk/castelul-corvin/), the Dacian side sits at [Sarmizegetusa in the mountains](https://mariuscomper.uk/sarmizegetusa-mituri/), the Apuseni gold sits at [Roșia Montană](https://mariuscomper.uk/rosia-montana/), the bridge sits at [Drobeta](https://mariuscomper.uk/podul-lui-traian/), and the spiral of the wars sits at the [Column](https://mariuscomper.uk/columna-lui-traian/). From Scaurianus's furrow to the gates blocked with tomb lids: the colony was a province capital, a 4th-century refuge, a stone quarry, the village Britonia — and today it is a lesson in reading stone. Two Sarmizegetusas. One in the mountains, one in the plain. The one in the plain is called Rome. ## Pick the colony's age Jump to the year that interests you; the 3D section shows the city of that time. ## How do you read a city from two ages? Start with what stands in place: the 600 by 500 metre curtain wall with its 33 hectares, the marble forum, the 88 by 69 amphitheatre with its channel, the sacred-area temples, the necropoleis on either side of the imperial road. Then set the years in pairs: 108–110 the founding with its unknown year, 124–128 the forum's completion after the coins, 158 the amphitheatre repairs after the tiles, 271–275 the withdrawal, the 4th century the blocked gates, 1315 Britonia, 1832 the mosaic, 1924 Daicoviciu, 1973 the resumption, 1982 the museum, 2024 UNESCO, 2025 the prize. Don't forget the sums: some 20,000 inhabitants (estimates run from 15,000 to 30,000, depending on the author), 5,000 seats in the amphitheatre, 12 entrances to the stands, the 66 by 47 arena, 4-metre colonnades, a 5-metre wall. And don't forget the brackets: missing inscription letters are restored in square brackets, and what sits in brackets is the archaeologist's hand, not the stone's. For class: draw the city in three strips: the colony (108–110), the arena (the games), the fortress (the 4th century). Mark the forum, the amphitheatre, the walls and the blocked gates. Which phase is missing between the 5th century and 1315 — and what question would you ask to fill it? ## Who walled, who prayed, who dug? Trajan remains the far-off authority and never showed at the founding — this page does not hand him the plough; the foundation is tied epigraphically to Scaurianus, with the 108–110 and 109–110 ranges and no exact year. The colonists were veterans and civilians from all over the empire, Palmyrenes with Malagbel included; the decurions judged in the forum, and the gladiators prayed to Nemesis at the arena gate. Hadrian completed the forum and — perhaps — completed the name; and Aurelian signed the 271–275 withdrawal, with no “departure day” at Ulpia. Between the wallers and us stand those who read: Mezerzius from around 1495, Marsigli with the amphitheatre map, Ackner with the 8.15 by 6.45 metre mosaic of 1832, Mommsen with the lost inscriptions, Király with Téglás at the temples, Constantin Daicoviciu from 1924, Floca in 1936 and again in 1948 at Hobița, the 1973 team with Alicu and Piso, Diaconescu with the 3D reconstruction — and today's German team, with the new temple. And the city died and rose twice: once when the arena became a fortress, and once in 2024, when it became world heritage. For class: lay the people on a timeline: 108–110, 124–128, 158, 222–235, 271–275, the 4th century, 1315, 1495, 1775, 1832, 1880, 1921, 1924, 1936, 1948, 1973, 1982, 2024, 2025. Where does the German team go on the line — at walling or at reading? What question would you ask to place it right? ## What is legend and what is record? Ulpia keeps its legends between forum and arena — and this page grades them, as at Argeș, Hunedoara, Voroneț, Peleș, Constanța and Adamclisi: the capital is half, Hadrian is unproven, Trajan with the plough is legend, Densuș-from-the-amphitheatre is legend, Aramaic is unproven, first-only-biggest are banned, “Regia yes, Ulpia no” is overturned. None of it is presented here as fact without paper. As for the name: the meaning of Sarmizegetusa stays unknown, and “the fortress on the rock” is a mere hypothesis — this page keeps it in quotes. And as for the capital: Ulpia was the symbolic, fiscal-administrative and religious capital of Dacia, though the governor usually resided at Apulum, in Macrea's words. The table grades them in short. The legends and facts of Ulpia, graded | Claim | Grade | Why | It was the capital of Roman Dacia | Half | symbolic and fiscal; Apulum, military | The Sarmizegetusa name came from Hadrian | Unproven | new epigraphy says: from the start | The foundation stone is Trajan's own work | Legend | Scaurianus, in his name | The Densuș church is from the amphitheatre stones | Legend | spolia yes, origin no | Aramaic was spoken here too | Unproven | Latin and Greek; Aramaic, at Tibiscum | It is the first, the only, the biggest | Banned | no comparison set | Regia is UNESCO, Ulpia is out | Overturned | both inscribed since 2024 For class: take something heard about Ulpia (“Trajan walled the whole city”, “Densuș is from the amphitheatre”) and grade it as in the table: museum fact, dig fact, half, overturned, legend, unproven or banned. What question would you ask to check it? ## Eight claims: true or false? Choose. The answers check against the page's sentences. - The Roman city and the Dacian capital shared the name Sarmizegetusa. - Trajan personally laid the colony's foundation stone. - After the Roman withdrawal, the amphitheatre gates were blocked with funerary pieces. - Aramaic inscriptions were found at Ulpia. - The Densuș church reuses Roman pieces in its walls. - The colony's exact founding year is known. - Today both Sarmizegetusas are UNESCO world heritage. - The meaning of the name Sarmizegetusa is known for certain. Send it on ## How we worked and what we don't know This page rests on the sources verified for this pass: the INP/CIMEC page, the MCDR Deva section with Gică Băeștean's text, the Princeton Encyclopedia, the UNESCO “Frontiers of the Roman Empire – Dacia” file with its maps, the RAN 91063.01 record, Cupcea's veteran study, the MNIR page on the Palmyrene temple, the 2025 PUZCP register — plus Alicu's amphitheatre monograph, through verified quotes. Years are told from records and the museum, not measured on the ground; legends are sorted by layer: museum, dig, press, legend, record. The 3D section is schematic: the amphitheatre shows its two states, arena and fortress, not after the architect's survey, and the plain is seen, not measured. What we don't know yet: the exact founding year, who added the name and when, the name's meaning, the camp beneath and its legion, the first colonists by name, what the pegma looked like, the schola's purpose, where Densuș's stones came from, the withdrawal day and the 5th–13th centuries between the last building and Britonia. When new records come, the page changes. ## Sources - [CIMEC: Ulpia Traiana Sarmizegetusa](https://cimec.ro/id-01-arheologie/situri-arheologice-22/ulpia-traiana-sarmizegetusa/) — the 108–110 founding, Scaurianus, the forum, the amphitheatre, the museum, the name hypothesis. - [MCDR Deva: Sarmizegetusa (Gică Băeștean's text)](https://mcdr.ro/index.php/en/sectii-si-compartimente-3/sectia-de-arheologie-sarmizegetusa-2) — the plough sestertius, the forum room by room, the pegma, the fortress, the spolia, the territorium. - [Princeton: Ulpia Traiana](https://www.perseus.tufts.edu/hopper/text?doc=PECS+ulpia-traiana) — 32.5 hectares, the wall, the 88 by 69 amphitheatre, the temples, the village over the city. - [UNESCO: the maps of file 1718](https://whc.unesco.org/en/list/1718/maps/) — component 1718-017 Ulpia and 1718-272 Regia. - [MNIT: the 7 August 2024 release](https://mnit.ro/en/-/2024/08/07/bunul-cultural-serial-frontierele-imperiului-roman-dacia-fre-dacia-a-fost-inscris-pe-lista-patrimoniului-mondial-unesco/) — the 27 July 2024 inscription, at New Delhi. - [RAN 91063.01](https://ran.cimec.ro/sel.asp?codran=91063.01&lang=EN) — the village's eastern ground, the codes, the protected area. - [Cupcea: Veteran settlement](https://hiphi.ubbcluj.ro/written_heritage/G.%20Cupcea,%20Veteran%20settlement.pdf) — barely two first-wave veterans, IIII Flavia, the camp as hypothesis. - [MNIR: the Palmyrene gods' temple](https://www.mnir.ro/colonia-ulpia-traiana-augusta-dacica-sarmizegetusa-templul-zeilor-palmireni/) — the Palmyrene community, the temple west of the walls. - [Judaism and Rome: the foundation inscription record](https://heurist.huma-num.fr/judaism_and_rome/tpl/public-record/120) — the bracketed text, the Latin language, the Wolff reconstruction. - Cover photo: Raimond Spekking, via Wikimedia Commons — the other images are credited under each photograph. Ulpia Traiana Sarmizegetusa: the 108–110 founding, the marble forum, the arena turned fortress and the graded legends. Text: Marius Comper. [Method](#metoda) · [Sources](#surse) · [mariuscomper.uk](https://mariuscomper.uk/)